Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Aug 2.
Published in final edited form as: Neuron. 2017 Jul 20;95(3):673–682.e4. doi: 10.1016/j.neuron.2017.06.045

Parallel inhibitory and excitatory trigemino-facial feedback circuitry for reflexive vibrissa movement

Marie-Andrée Bellavance 1,*, Jun Takatoh 2,*, Jinghao Lu 2, Maxime Demers 1, David Kleinfeld 3,4,5,†, Fan Wang 2,†, Martin Deschênes 1,†
PMCID: PMC5845798  NIHMSID: NIHMS947312  PMID: 28735746

Summary

Animals employ active touch to optimize the acuity of their tactile sensors. Prior experimental results and models lead to the hypothesis that sensory inputs are used in a recurrent manner to tune the position of the sensors. A combination of electrophysiology, intersectional genetic viral labeling and manipulation, and classical tracing, allowed us to identify second-order sensorimotor loops that control vibrissa movements by rodents. Facial motoneurons that drive intrinsic muscles to protract the vibrissae receive a short latency inhibitory input, followed by synaptic excitation, from neurons located in the oralis division of the trigeminal sensory complex. In contrast, motoneurons that retract the mystacial pad, and indirectly retract the vibrissae, receive only excitatory input from interpolaris cells that further project to the thalamus. Silencing this feedback alters retraction. The observed pull-push circuit for motion control at the lowest-level sensorimotor loop provides a mechanism for the rapid modulation of vibrissa touch during exploration of peri-personal space.

Keywords: brainstem, facial nucleus, intersectional labeling, spinal trigeminal nuclei, touch, whisking

INTRODUCTION

The most basic unit of motor action is a reflex arc. A sensory organ is activated by a stimulus and drives a motoneuron that leads to a response in an effector cell, usually a muscle. Reflex arcs are typically located in the hindbrain or spinal cord and mediate fast responses through positive or negative feedback loops. While the reflex arc can be formed by a direct connection from a sensory afferent to a motoneuron, frequently reflex arcs include a second-order neuron, or interneuron, between the sensory and motor neuron. The second-order neuron may play two roles: first, as a gate to modulate the effectiveness of the reflex arc, and second, as a relay to convey information to the brain or other parts of the body.

Reflex arcs are fundamental to touch that is mediated by the vibrissa system of rodents. Rhythmic motion of the vibrissae is driven by a central oscillator that can operate independently of direct sensory feedback (Welker, 1964; Gao et al., 2001; Berg and Kleinfeld, 2003; Moore et al., 2013). The effect of vibrissa contact on whisking kinematics has been studied in freely moving and head-restrained rats (Sachdev et al., 2003; Mitchinson et al., 2007; Grant et al., 2008; Deutsch et al., 2012; Hires et al., 2013; Voigts et al., 2015). The results of these studies are not clearly consistent. Upon contact with an object, vibrissae on the side of contact may rapidly and transiently protract (Sachdev et al., 2003), retract (Mitchinson et al., 2007), or exhibit a pump-like motion of retraction followed by protraction, denoted as a touch-induced pump by Deutsch et al. (2012) (Fig. 1A). As discussed by Sherman et al. (2013), neuro-reflexive motion can occur through parallel pathways in the brainstem (Fig. 1B). Reflexive retraction may occur by transient inhibition of the intrinsic motoneurons and/or by excitation of motoneurons that drive the extrinsic retractor muscles, i.e., maxolabialis and nasolabialis. Reflexive protraction may occur by transient excitation of the intrinsic motoneurons and/or by excitation of motoneurons that drive the extrinsic protraction muscle nasolabialis profundus.

Figure 1. Basic behavior and vibrissae mechanics.

Figure 1

Open in a new tab

(A) Vibrissa contact leads to a pump-like motion, with a transient dip in the protraction of the vibrissae and an elongation of the interval of protraction. The upper illustration shows the set-up to observe contact while a head-fixed rat whisks, while the lower image shows the angular dependence of the vibrissae during whisking is air versus touch to a pole. Data from Deutsch et al. (2012).

(B) Cartoon of the muscles that control protraction and retraction of the vibrissa and may be involved in the reflex. Activation of the intrinsic muscles are the major drive of protraction. Activation of the extrinsic muscle nasolabialis profundus pulls the mystacial pad rostral and thus shifts the vibrissae toward protraction, while activation of the extrinsic muscles nasolabialis and maxillolabialis pull the pad caudal and thus shifts the vibrissae toward retraction. The extrinsic muscle fibers (emfs) run underneath the skin between rows of vibrissae. Illustration by J. Kuhl.

Independent of the details of the motion, the lack of direct projections from primary vibrissa afferents to facial motoneurons (Hattox et al., 2002; Takatoh et al., 2013) implies that sensory feedback involves second-order, or higher-order, neurons. In fact, experiments in brainstem slice indicate that sensorimotor feedback is mediated by a disynaptic pathway that lies in the hindbrain (Nguyen and Kleinfeld, 2005). This circuit involves premotor neurons located in the oralis (SpVO) and rostral interpolaris (SpVIr) subnuclei of the spinal trigeminal complex (Pinganaud et al. 1999; Takatoh et al., 2013; Sreenivasan et al, 2015). Transsynaptic labeling further revealed a differential distribution of trigeminal premotor neurons. Those that innervate the intrinsic protractor motoneurons are principally located in subnucleus SpVO, while premotor neurons that drive the extrinsic retractor motoneurons in the mystacial pad are preferentially clustered in subnucleus SpVIr (Takatoh et al., 2013; Sreenivasan et al., 2015). Thus the known anatomy can support a mixed inhibitory and excitatory response (Deutsch et al., 2012; Sherman et al. 2013). Yet the specifics of this circuitry and the potential involvement of different muscle groups in the reflex arc (Fig. 1B) remains an open issue.

Here we used intracellular recording, virus based labeling methods, including intersectional genetic viral labeling, for tract tracing and neuronal activation and inactivation, as well as classical tracers, to decipher the circuitry involved in reflex control of facial motoneurons. We further extend past studies to include muscle nasolabialis profundus (Hill et al., 2008).

RESULTS

Anatomical background

The facial nucleus is the sole output station that controls the activation of vibrissa muscles. It contains only alpha motoneurons (Moore et al, 2015), no local circuit cells, and its motoneurons do not give off intranuclear axon collaterals (Friauf, 1986). Several studies have shown that the rodent facial musculature is anatomically mapped within the facial motor nuclear complex (Watson et al., 1982; Ashwell, 1982; Hinrichsen and Watson, 1984; Komiyama et al., 1984; Furutani et al., 2004; Klein and Rhoades, 2004; Takatoh et al., 2013; Sreenivasan et al., 2015; Deschênes et al., 2016). Motoneurons that innervate the intrinsic vibrissa muscles are located in the ventral lateral part of the nucleus, the extrinsic retractor muscles nasolabialis and maxillolabialis are represented dorsolaterally, and the extrinsic protractor muscle nasolabialis profundus is represented at the lateral most edge of the nucleus.

As a means to map premotor input to the extrinsic protractor muscle, ΔG-Rabies (Wickersham et al., 2007) was injected into muscle nasolabialis profundus (5 juvenile mice). We observed retrograde labeling at the extreme lateral edge of the facial nucleus, transsynaptic labeling in the parafacial region and in the preBotzinger complex, but virtually no transsynaptic labeling in trigeminal sensory nuclei (3 cells in 5 mice; Fig. 2). We conclude that nasolabialis profundus motoneurons are not likely involved in short-latency reflex motion to whisker contact.

Figure 2. Absence of innervation of trigeminal nuclei by the extrinsic protractor muscles.

Figure 2

Open in a new tab

(A–D) Injection of ΔG-rabies-mCherry in the tip of the snout (panel A) leads to retrograde labeling in the lateral most part of the facial nucleus (panel B), and transsynaptic labeling in the parafacial region (panel C) and the preBötzinger complex (panel D). Note the absence of labeled cells in the oralis (SpVO) and interpolaris (SpVIr) divisions of the spinal trigeminal sensory complex. Abbreviations: Amb, ambiguous nucleus; FN, facial nucleus; PF, parafacial region; pBötC, preBötzinger complex. Lower-case letters in panel A indicate the corresponding vibrissa rows.

Response of facial motoneurons to vibrissa deflection in anesthetized rats

In a second series of experiments (8 rats) we examined synaptic events evoked in facial motoneurons by passive deflection of the vibrissae (Fig. 3). Motoneurons were identified as intrinsic motoneurons on the basis of single-vibrissa protractions elicited by the firing of action potentials during intracellular current injection (20 cells). Otherwise, motoneurons were considered as extrinsic motoneurons, which include motoneurons whose firing led to retraction of 2–6 vibrissae (4 cells), and motoneurons that displayed respiratory modulation (3 cells). The other motoneurons remained uncharacterized (5 cells).

Figure 3. Intracellular evidence for sensory feedback in facial motoneurons following deflection of the vibrissae.

Figure 3

Open in a new tab

(A, B) In intrinsic motoneurons vibrissa deflection induces a short-latency hyperpolarization at an average latency of 3.5 ± 0.4 ms. In some intrinsic motoneurons, shown as overlaid traces in panel B, the early hyperpolarization was interrupted or followed by depolarizing events that triggered an action potential (arrows; action potentials are cropped).

(C, D) In intrinsic motoneurons, the hyperpolarization triggered by air puff deflection of the vibrissae reversed in sign upon hyperpolarizing current injection (panel C), and became depolarizing shortly after cell impalement with a KCl-filled pipette (panel D).

(E) In pad retractor motoneurons, passive vibrissa deflection induces a prompt excitation. The ten superimposed traces in this panel are from a motoneuron that drove retraction of vibrissae E1, E2, E3, D1, D2, and D3.

(F) The time course of sensory-evoked responses in an intrinsic motoneuron and a pad retractor motoneuron with that of the local field potential (LFP) recorded in the subnucleus SpVO. Each trace is the average of 35 responses. Note the very short delay, about 0.5 ms, between the onset of the LFP (dashed line) and the onset of intracellular events.

(G) Traces show how the latency of protraction after a motoneuronal spike was measured. In this example, spikes were triggered by current injection in a D1 motoneuron, and a piezoelectric film was used to monitor displacement of vibrissa D1 (red trace).

(H) Spike-triggered average protraction of vibrissa D1 (34 events).

In ketamine/xylazine anesthetized rats, facial motoneurons had no spontaneous activity and air-puff deflection of the vibrissae did not elicit extracellular spiking activity. When cells were recorded intracellularly, we observed a short-latency hyperpolarization in all intrinsic motoneurons (20 cells) upon deflection of the vibrissae (Fig. 3A), irrespective of whether air puffs deflected the vibrissa driven by the cell or the adjacent vibrissae. Air puffs directed toward the tip of the nose did not elicit any response. The hyperpolarization had a mean onset latency of 3.5 ± 0.4 ms, and was often followed by depolarizing events that occasionally led to spike discharge (arrows, Fig. 3B). The hyperpolarization reversed polarity at about −70 mV (Fig. 3C), and became rapidly depolarizing, i.e., within 1 minute, after impalement with a KCl-filled pipette (5 cells; Fig. 3D). Thus, this early event was identified as a chloride-mediated inhibitory postsynaptic potential (IPSP).

In pad retractor motoneurons (4 cells), deflection of a vibrissa induced a depolarization at a mean latency of 3.3 ± 0.2 ms (4 cells; Fig. 3E); no response was observed in the other motoneurons (8 cells). The data of Figure 3F illustrate the time course of vibrissa-evoked responses in facial motoneurons together with the local field potential recorded in subnucleus SpVO. One can appreciate the fast onset of motoneuronal responses, which lags the onset of the field potential by a mere 0.5 ms.

Assessment of reflex latency

The latency of a reflex is the sum of the time delay for motoneuron activation after stimulus onset, plus the time delay for muscle contraction after motoneuron firing. From the above intracellular recordings, motoneurons were activated/inhibited about 3.5 ms after stimulus onset. To measure the latency of vibrissa motion induced by the spiking of motoneurons, vibrissae were observed under a stereomicroscope to detect movements induced by cell firing driven by current injection. Once protraction of a single vibrissa was detected, a piezoelectric film was placed in contact with the rostral edge of the vibrissa. This allowed an accurate measurement of the time delay between the firing of an action potential and initiation of vibrissa protraction (Fig. 3G,H). The mean latency of protractions was 7.5 ± 0.4 ms (10 cells; 1293 spikes), which is within the range of latencies, i.e., 7.6 ± 2.5 ms, reported by video tracking (Herfst and Brecht, 2008). Together these results indicate that the shortest reflex latency in the vibrissa-trigemino-facial loop is in the range of 11 to 13 ms, which is in accord with the reflex latency of 13 ms estimated for freely exploring rats (Mitchinson et al., 2007).

Anatomical substrate for reflex excitation and inhibition in rats

The short latency of vibrissa-evoked responses in facial motoneurons indicates that sensory inputs must arise from the sensory trigeminal complex. Transsynaptic labeling studies in mice showed that the subnucleus SpVO contains both glutamatergic and GABAergic/glycinergic premotor neurons that target motoneurons that drive the intrinsic vibrissa muscles. In contrast, subnucleus SpVIr contains a single population of glutamatergic premotor neurons that target motoneurons that drive the extrinsic retractor muscles (Takatoh et al., 2013; Sreenivasan et al., 2015). We thus investigated the origin of the trigemino-facial excitatory and inhibitory inputs that target the intrinsic and pad retractor motoneurons in rats. Fluorogold was injected iontophoretically in the ventral lateral facial nucleus, and parasagittal sections of the trigeminal sensory nuclei were processed for in situ hybridization using a VGAT probe. Three injections were restricted to the ventral lateral part of the facial nucleus (Fig. 4A) with no spread of the tracer in the dorsolateral sector where retractor muscles are represented. In these three cases we found retrogradely labeled cells nearly exclusively in the oralis nucleus (Fig. 4A), where 24 % (159 out of 652) of labeled cells expressed the VGAT transcript (Fig. 4B). Assuming that oralis neurons are glutamatergic or GABAergic/glycinergic (Ge et al., 2014; Avedaño et al., 2005; Takatoh et al., 2013), one can infer that oralis premotor neurons provide both glutamatergic and GABAergic/glycinergic inputs to motoneurons that drive the intrinsic vibrissa muscles.

Figure 4. Anatomical substrate for the trigemino-facial reflex.

Figure 4

Open in a new tab

(A–B) Fluorogold was injected in the ventral lateral facial nucleus. The arrowhead indicates the location of a cluster of retrogradely labeled cells in subnucleus SpVO. The dashed lines delineate subnucleus SpVO. In situ hybridization was used to label subnucleus SpVO neurons that express VGAT (arrowheads in panel B).

(C–F) Interpolaris neurons that project to Po thalamus were selectively labeled by injecting RG-LV-hSyn-Cre in Po thalamus, and AAV.2/1.hSyn.ChR2.EYFP.WPRE in subnucleus SpVIr. The framed area in panel C is enlarged in panel D. Anterograde labeling of interpolaris axons reveals terminal fields in Po thalamus and the ventral part of zona incerta (panel D), as well as in the dorsal lateral division of the facial nucleus (panel E). The dashed line in panel E indicates the dorsal border of the facial nucleus. The preferential distribution of subnucleus SpVIr terminals in the dorsal lateral sector of the facial nucleus is illustrated in the 3D map in panel F (a, anterior; d, dorsal). Sagittal sections (panels C–F) were counterstained for cytochrome oxidase. Abbreviations: PrV, principal trigeminal nucleus; SpVO, oralis subnucleus of the spinal trigeminal complex; SpVIr, rostral division of the interpolaris subnucleus of the spinal trigeminal complex; SpVIc, caudal division of the interpolaris subnucleus of the spinal trigeminal complex; SpVM, muralis sector of the spinal trigeminal complex; SpVC, caudal subnucleus of the spinal trigeminal complex; ZIv, ventral division of zona incerta.

It was previously reported that trigeminal premotor neurons that drive the extrinsic retractor motoneurons preferentially cluster in SpVIr (Sreenivasan et al, 2015). We thus examined whether interpolaris cells that project to the facial nucleus also project to the posterior group (Po) of the thalamus. We used a conditional labeling strategy, which consists of injecting a retrograde lentivirus that express Cre in Po thalamus (RG-LV-hSyn-Cre) and an AAV Flex-eGFP vector in SpVIr (Fig. 4C – E). This approach revealed a collateral projection to the facial nucleus from interpolaris cells that project to Po thalamus (Fig. 4F). The three dimensional map of Figure 4G shows that the vast majority of labeled terminals are located in the dorsal lateral sector of the nucleus, where pad retractor motoneurons are clustered. This result is direct evidence that reflex activation of pad retractor motoneurons is mediated at least in part by glutamatergic interpolaris cells that project to Po thalamus.

Anatomical substrate for trigemino-facial reflex in mice

A prior study in mice revealed that subnucleus SpVO provides both glutamatergic and GABAergic/glycinergic inputs to the intrinsic protractor motoneurons, while glutamatergic SpVIr neurons innervate the extrinsic retractor motoneurons (Takatoh et al., 2013). We thus examined whether interpolaris cells that innervate pad retractor motoneurons also project to Po thalamus in mice. Motivated by a past study (Chiala et al., 1992), we found that parvalbumin (Pv) is expressed in excitatory (glutamatergic) SpVIr neurons. To determine whether Pv+ SpVIr cells represent the premotor neurons that project to both Po thalamus and extrinsic facial motoneurons, we employed a split-Cre system (Wang et al., 2012; Stanek et al., 2016). Specifically, we generated Pv-CreN knockin mice, and injected retrograde lentivirus that expressies CreC (RG-LV-CreC) in Po in Pv-CreN; Ai14 (Cre-dependent tdTomato reporter) double transgenic mice. In this way, Pv+ SpVIr cells projecting to Po were labeled with tdTomato. In situ hybridization (3 mice) demonstrated that the vast majority (97 ± 1 %) of tdTomato labeled neurons are vGlut2 positive (Fig. 5A,B). We next injected RG-LV-CreC into Po thalamus of Pv-CreN mice (4 mice; Fig. 5C), and AAV encoding Cre-dependent GFP was injected into SpVIr. This approach ensures that only Pv-positive SpVIr neurons projecting to Po thalamus are labeled in each mouse. We found that, just like in rats, GFP-labeled SpVIr neurons that project to Po thalamus give off collaterals to the dorsolateral sector of the facial motor nucleus (Fig. 5D,E). Optogenetic stimulation of Po-projecting SpVIr neurons that express channelrhodopsin further confirms that the SpVIr-Po-facial dual projection activates muscle nasolabialis (6 mice; Fig. 5 F,G).

Figure 5. Characterization of subnucleus SpVIr neurons that project to Po thalamus and are involved in the trigemino-facial reflex.

Figure 5

Open in a new tab

These data make use of a split-Cre strategy (Wang et al., 2012; Stanek et al., 2016). The starting substrate in all cases are mice that are engineered to express CreN in neurons that naturally express parvalbumin. Retrograde lentivirus pseudotyped with FuGB2-coat that encodes CreC (RG-LV-CreC) was injected into Po thalamus. Thus functional Cre is reconstituted in PV-positive neurons that project to Po thalamus.

(A) Schematic of experiment to confirm that PV-neurons in subnucleus SpVIr that further project to Po thalamus will form excitatory synapses. PV-CreN mice were crossed to tdThomato reporter mice so that all neurons contain the Cre-dependent tdTomato (Pv-CreN;Ai14). Here, the functional Cre leads to expression of tdTomato in PV-positive neurons that project to Po thalamus.

(B) In situ hybridization performed on mice prepared as in panel A shows that vGlut2 (pseudo-colored red) is co-expressed in neurons that stain positive for td-tomato (anti-RFP pseudo colored green). Coronal section; counter staining with DAPI.

(C) Schematic of experiment to confirm that PV-neurons in subnucleus SpVIr that further project to Po thalamus also project to the facial motor nucleus. AAV that encodes Cre-dependent GFP (AAV-hSyn-Flex-GFP) was injected into subnucleus SpVIr at the same time that RG-LV-CreC was injected into Po thalamus in PV-CreN mice. Only PV-positive SpVIr neurons that project to PO-thalamus will express GFP, and the axonal collaterals of labeled neurons in FN can be visualized.

(D) Low and high magnification views of terminals (green) in Po-thalamus for mice prepared as in panel C. Arrows point to labeled terminals. Coronal section; counter staining with NeuroTrace blue.

(E) Terminals (green) in FN for mice prepared as in panel C. Coronal section innervating the region where extrinsic motoneurons are located; counter staining with Neurotrace blue.

(F) Schematic of experiment to confirm that PV-neurons in SpVIr that further project to Po thalamus can drive motoneurons for the extrinsic muscles in the mystcial pad (Fig 1B). PV-CreN mice were crossed to channelrhodopsin (ChR2) reporter mice so that all neurons contain Cre-dependent (Pv-CreN;Ai320). Here, the functional Cre leads to expression of ChR2 in PV-positive neurons that project to Po thalamus. A cannula for an optical fiber was implanted into subnucleus SpVIr and electromyogram (EMG) electrodes were implanted in the extrinsic muscles (Berg and Kleinfeld 2003).

(G) Data from mice, lightly anesthetized with ketamine/xylazine, prepared as in panel G. Light pulses at 440 nm, 20 ms wide, were delivered at 500 ms internals. The ChR2 mice showed retraction muscle activation upon light stimulation while control animals (without RG-LV-CreC injection) did not show any responses to the light stimulation.

Silencing SpVIr premotor neurons prolongs vibrissa contact time

Lastly, we examined whether silencing the SpVIr-facial projection affects the kinematics of whisking upon vibrissa contact in naive, head-restrained mice. This projection was inactivated unilaterally by injecting RG-LV-CreC into the Po thalamus of Pv-CreN animals and AAV encoding Cre-dependent TeLC-GFP into SpVIr (6 mice; Fig. 6A,B). Contact-induced transient vibrissa reaction was assessed by a change in vibrissa curvature (Fig. 6C). In the naive mice used for this manipulation, retraction was observed in 3 % and 7 % of the cases on the normal and silenced side, respectively. The profound effect of silencing the SpVIr-facial projection was to significantly prolong contact on the silenced side (Fig. 6D; see marginal on top), consistent with a lack of involvement of active retraction through the extrinsic retractor muscles (Fig. 1B). On the other hand, and consistent with the absence of retractor muscles activity at the onset of protraction, the mechanical impulse (Fig. 6D; insert) applied during the initial phase of contact remains unaffected by silencing the SpVIr-facial projection (Fig. 6D; see marginal on right).

Figure 6. Silencing of subnucleus SpVIr neurons that project to Po thalamus alters the trigemino-facial reflex.

Figure 6

Open in a new tab

(A) Schematic of experiment. RG-LV-CreC is injected into Po thalamus in PV-CreN mice so that functional Cre is reconstituted in PV-positive neurons that project to Po thalamus. Concurrent with the RG-LV-CreC injections, AAV that encodes Cre-dependent expression of tetanus toxin light chain (TeLC) linked to GFP (AAV-hSyn-Flex-TeLC-GFP) was injected into subnucleus SpVIr. Thus, only PV-positive SpVIr neurons that project to both PO-thalamus, as well as send collaterals to the FN, will have their transmission silenced to tetanus toxin

(B) Post hoc histology shows expression of TeLC-GFP (pseudo colored green) in SpVIr cells. Coronal section; counter staining with NeuroTrace blue.

(C) Changes in the curvature, k, of the vibrissa upon contact with an edge (red); contact force is proportional to the curvature. At high forces a transient drop, or dip, occurs (see also Fig 1A).

(D) Scatter plot of the impulse upon contact versus the width of the contact interval for contact on the intact side versus the side with feedback to the extrinsic retractor muscle, i.e., PV-positive SpVIr neurons that project to both PO-thalamus and to the extrinsic facial motoneurons, silenced. The impulse is computed as the curvature, which serves as a surrogate for the force at the base of the vibrissa, integrated over the onset of the contact (Inset), i.e., Impulse=∫0First peakdt κ(t). The interval is taken as the full period of contact. The computed impulse is the same for intact and silenced sides, as shown by the marginal (right side; p > 0.1; two-tailed KS test), but the widths are different (top, p < 10−6; two-tailed KS test).

DISCUSSION

We have completed the wiring diagram of the shortest sensorimotor loops for reflex motion of the vibrissae upon contact with an object (Fig. 7). Primary vibrissa afferents excite GABAergic/glycinergic oralis neurons as well as glutamatergic cells in the oralis and interpolaris subnuclei. In turn, both excitatory and inhibitory oralis cells project to the motoneurons for intrinsic muscles, and activate or suppress, respectively, motoneuron output for the intrinsic muscles that drive vibrissae protraction (Figs. 3 and 4). In contrast, Po-projecting interpolaris cells excite the motoneurons for extrinsic retractors by means of axon collaterals (Figs. 3–6). Thus there are two pathways that can account for fast vibrissa retraction induced by vibrissa contact in prior behavioral studies (Mitchinson et al., 2007; Deutsch et al., 2012), i.e., inhibition of the intrinsic motoneurons, and excitation of the extrinsic retractor motoneurons. In contrast to the case for retraction, only one pathway can account for reflexive protraction reported by Nguyen and Kleinfeld (2005) and the touch-induced pumps by Deutsch et al. (2012). In toto, these pathways allow both kinds of reflexes at the appropriate latencies, and their architecture facilitates top-down control of their execution and gain.

Figure 7. Schematic diagram of first-order feedback loops involved in reflex motion of the vibrissae.

Figure 7

Open in a new tab

Intrinsic motoneurons in the ventral lateral sector of the facial nucleus receive both excitatory and inhibitory inputs from subnucleus SpVO, whereas nasolabialis and maxillolabialis motoneurons in the dorsal lateral sector receive excitatory input from subnucleus SpVIr neurons that also project to Po thalamus. Abbreviations: principal trigeminal nucleus (PrV), spinal trigeminal subnuclei oralis (SpVO), interpolaris (SpVI), muralis (SpVM), caudalis (SpVC), muscles nasolabialis (NL) and maxillolabialis (ML) that retract the pad, muscle nasolabialis profundus (NLP) that protracts the pad, and muscle deflector nasi (DN) that deflects the nose.

Nguyen and Kleinfeld (2005) found exclusively short-latency reflexive excitation of intrinsic and/or nasolabialis motoneurons after stimulation of primary trigeminal afferents. Data were obtained from recording in horizontal brainstem slices in young rats, and from electromyographic data in anesthetized adult animals. Two factors likely explain why vibrissa-evoked inhibition of intrinsic motoneurons, or related muscular suppression, was unnoticed in this earlier study. First, the effect of inhibition of the intrinsic motoneurons cannot be detected with electromyographic recordings in anesthetized animals because motoneurons that innervate the intrinsic muscles are already silent. Second, retrograde labeling shows that the inhibitory input to intrinsic motoneurons arises from oralis cells located dorsally relative to the depth of the facial nucleus. Thus, in horizontal slices that include the facial nucleus, most of the inhibitory trigemino-facial connections were likely severed, leaving only the excitatory response. Other studies have addressed feedback connectivity from the trigeminus to the facial nucleus in the adult rodent. Sreenivasan et al. (2015) identified a pathway that corresponds to neurons in subnucleus SpVIr that project to Po thalamus and the extrinsic muscles (Fig. 7). Further, a projection from the muralis division of the trigeminal sensory complex to the facial nucleus has been reported (Matthews et al., 2015). This projection was proposed to be part of a reflex pathway. However, recent anatomical data (Tonomura et al 2015) and ongoing electrophysiological experiments (Callado-Pérez et al., 2016) do not support vibrissa touch signals as afferent input to muralis neurons.

Context-dependence of the trigemino-facial reflex

Inhibitory and excitatory postsynaptic potentials evoked in intrinsic motoneurons by air-puff vibrissa deflection occurred at nearly the same latency. Yet the dominant effect was inhibition (Fig. 3B). As suggested by the rapid reversal of IPSPs after impalement with a chloride-containing pipette, and recalling that dendrites of facial motoneurons are electrotonically long (Nguyen et al., 2004), the proximal location of inhibitory synapses likely explains the shunting of more distally located excitatory inputs. A critical issue, though, is the behavioral conditions under which inhibition or excitation prevails. Passive vibrissa deflection produced by an air puff is a useful means to uncover sensorimotor feedback loops involved in reflex control of facial motoneurons. However, this approach does not actually mimic the dynamics of reflex pathways engaged upon vibrissa contact during exploratory behaviors. Given the disynaptic nature of the reflex, its gain is likely modulated by motor strategies and the context of exploration, such as whether exploration occurs in a familiar versus unknown environment (Gordon et al., 2014). It was reported that when the vibrissae touch a non-attended surface they show more pronounced bending, suggesting that contact-induced vibrissa is related in some way to the current goals and focus of the animal (Mitchinson et al., 2007; Voigts et al., 2015). Further, reflex retraction is reported to occur only for a quarter of the epochs when the vibrissae contact an object in the head-restrained rat (Deutsch et al., 2012). Yet when the rats intended to explore an object, the probability of reflexive motion more than doubled (Sherman et al., 2017).

How can high-level input control a low-level, disynaptic vibrissa reflex? It is worth reminding that trigeminal sensory nuclei are replete with inhibitory circuits that operate at both the pre- and postsynaptic levels (Ide and Killackey, 1985; Bae et al., 2000, 2005; Furuta et al., 2008) These circuits can be modulated by top-down pathways as well as by cholinergic, serotonergic and noradrenergic modulatory systems (reviewed in Bosman et al., 2011). Just like in the spinal cord, where both presynaptic and postsynaptic inhibition gates cutaneous inputs during locomotion (Rossignol et al., 2006), similar mechanisms might control the gain and sign of reflex actions for all orofacial pathways mediated by trigeminal neurons that project to the facial nucleus. In particular, the results of modeling efforts (Sherman et al., 2013) imply that high-level feedback may change the relative contribution of intrinsic versus extrinsic muscles in the control of the pump-like motion of retraction followed by protraction upon touch.

EXPERIMENTAL PROCEDURES

Subjects

Experiments were carried out in 22 Long Evans female rats (250–350 g in mass), and in 5~8 weeks old adult mice (both male and female) according to the National Institutes of Health Guidelines.

Pv-CreN knock-in mice were generated by inserting a 2A sequence, CreN-Intein-N (Wang et al. 2012) and Frt-flanked PGK-neomycin resistance cassette immediately in front of the stop codon of parvalbumin coding sequence by homologous recombination. ES clones with appropriate homologous recombination were selected by Southern blotting. After the knock-in mouse line was established, the Frt-flanked PGK-neomycin resistance cassette was removed by crossing with transgenic mice expressing Flpo in all cells (Jackson Laboratory, Stock No: 007844).

The Ai14 reporter line, which carries a Cre-dependent tdTomato cassette (Jackson Laboratory Stock No: 007908), and the Ai32 reporter line, which carries a Cre-dependent ChR2-EYFP cassette (Jackson Laboratory Stock No: 012569), were crossed to generate Pv-CreN;Ai14 and Pv-CreN;Ai32 mice, respectively.

All experiments were approved by the Institutional Animal Care and Use Committee at Duke University, Laval University, and the University of California at San Diego.

Viruses

RG-LV-CreC and RG-LV-Cre were designed and produced as described (Stanek et al, 2016). AAV2/8-hSyn-Flex-TeLC-P2A-GFP was designed and produced as described (Zhang et al., 2015). ΔG-Rabies-mCherry was produced as described (Takatoh et al., 2013). AAV2/8-hSyn-DIO-GFP (8 × 1012 ifu/ml) was purchased from the University of North Carolina Vector Core. AAV2/1-EF1a-ChR2-EYFP-WPRE (6 × 1012 ifu/ml) was purchased from the University of Pennsylvania Vector Core).

Recording of facial motoneurons

Rats were anesthetized with ketamine (75 mg/kg) and xylazine (5 mg/kg), and body temperature was maintained at 37.5°C with a thermostatically controlled heating pad. Respiration was monitored with a cantilevered piezoelectric film (LDT1 028K; Measurement Specialities) resting on the rat’s abdomen just caudal to the torso. Facial motoneurons were recorded intracellularly with micropipettes (tip diameter, 1µm) filled with either 0.5 M potassium acetate, or 3 M potassium chloride. Vibrissae were deflected rostralward or caudalward with the use of 10 – 70 ms air-puffs. The time delay between the command voltage and the onset of vibrissa deflection was measured with a piezoelectric film positioned at the same distance from the puffer tip. A piezoelectric film was used to measure the latency of vibrissa protraction induced by motoneuron spiking. An insect pin was glued to the piezoelectric film and placed in contact with the rostral edge of the vibrissa. Although resonance of the sensor altered the apparent time course of vibrissa retraction, this allowed an accurate measurement of protraction onset. All signals were sampled at 10 kHz and logged on a computer using the Labchart acquisition system (AD Instruments).

Retrograde labeling and in situ hybridization

To determine the origin and neurotransmitter content of trigemino-facial neurons, we combined retrograde labeling with in situ hybridization in rats. Fluorogold, prepared 1% (w/v) in cacodylate buffer, pH 7.1, was delivered iontophoretically in the ventral lateral facial nucleus using micropipettes (tip diameter, 8 µm) and positive current pulses (250 nA; duration 2 s; half-duty cycle for 15 min).

After a survival of 3 days, rats were perfused with paraformaldehyde, 4% (w/v), and parasagittal sections of the brainstem were cut on a freezing microtome (50 µm thickness) and processed for in situ hybridization using a vesicular GABA transporter (VGAT) probe according to standard protocols (Takatoh et al., 2013). Sections were then immunostained with a rabbit anti-Fluorogold antibody (Fluorochrome LLC), and an anti-rabbit IgG conjugated to Alexa 488 (Jackson ImmunoResearch). Cell counts were made from stacks of confocal images scanned at 20× magnification. As a reference, oralis subnucleus of the spinal trigeminal complex was defined as the region situated between the rostral and caudal edges of the facial nucleus. A slide scanner (TissueScope™ CF, Huron Technologies) was used to acquire large-scale images that were imported in Neurolucida (MBF Bioscience) to map the location of cells and axonal terminals labeled by virus infection.

Virus injection for intersectional labeling

To label premotor neurons that innervate muscle nasolabialis profundus, we injected ΔG-Rabies-mCherry (1 µl) in the tip of the snout in five juvenile mice (P2–P6) that expressed the missing G complement in cholinergic neurons (Takatoh et al., 2013). Mice were perfused after a survival period of 5 days.

To determine whether facial premotor neurons in trigeminal nuclei also project to the thalamus we injected a pseudotyped lentivirus expressing CRE recombinase (RG-LV-hSyn-Cre) in the posterior group (Po) of the thalamus (300 nl) and AAV.2/1.hSyn.ChR2.EYFP.WPRE in the rostral part of the nucleus interpolaris (SpVIr) (100 nl). Rats were perfused after a survival of 12 days.

In both of the above cases, parasagittal sections of the brainstem were cut at 50 µm, immunostained with a rabbit anti-GFP antibody (1:1000; Novus Biological), a biotinylated anti-rabbit IgG (1:200; Vector Labs), the avidin/biotin complex (Vectastain ABC kit; Vector Labs), and the SG peroxidase substrate (Vector Labs). Sections were counterstained for cytochrome oxidase to outline trigeminal sensory nuclei as well as the facial nucleus. Confocal microscopy was employed to count the number of doubly labeled cells in experiments that combined retrograde labeling with in situ hybridization.

For intersectional labeling RG-LV-CreC (800 nl) was injected in Po thalamus of Pv-CreN mice, and AAV-hSyn-Flex-GFP (500 nl) was injected in the SpVIr. Viruses were delivered at 100 nl/min using a glass capillary connected to an UltraMicroPump controlled by a SYS-Micro4 Controller (World Precision Instruments). After a survival of over 3 weeks mice were perfused as above. In situ hybridization was performed as described (Liang et al., 2000; Takatoh et al., 2013) on 40 µm floating sections that were hybridized with either DIG-labeled vGAT probe or vGlut2 probe. Cell counts were made from stacks of confocal images scanned at 20-× magnification. Immunofluorescence was performed on 80 µm floating sections that were counter-stained for NeuroTrace (Molecular Probes, N21479). Sections were imaged with a confocal microscope (LSM700; Carl Zeiss).

Electromyogram and optogenetic activation

Pv-CreN;Ai32 mice were injected with RG-LV-CreC into Po thalamus. After viral injection and optic cannula implantation, EMG electrodes were placed in the nasolabialis muscle as described previously (Berg and Kleinfeld 2003). For optogenetic activation experiments, mice were lightly anesthetized with ketamine and xylazine and held in a custom-made head-fix apparatus. Mice were connected to a blue laser (473 nm, OptoEngine LLC) through a fiber optic (200 µm core diameter, Thorlabs) with ferrule connector (Thorlabs) that was implanted above the SpVIr and secured with cyanoacrylate and Meta-bond (Parkell). Twenty ms laser pulses were delivered at 500 ms interval.

Whisking-based touch

Pv-CreN mice were injected with RG-LV-CreC and AAV2/8-hSyn-Flex-TeLC-P2A-GFP into Po thalamus and subnucleus SpVIr, respectively. Behavioral experiments were performed 14–21 d after virus injection. Mice were held in a custom-made head-fix apparatus using the pre-attached head plate and the mouse body was positioned into a 3D-printed body tube. Mice were acclimatized to head-fixation in the apparatus. One day before behavioral experiments, mice were anesthetized and all vibrissae except the C row vibrissae were trimmed. During the recording sessions, mice were head-restrained and presented with a vertical pole. All experiments were carried out in the dark under infrared illumination. Spontaneous touch behavior was captured with a high-speed camera (Basler) at 500 fps. Subnucleus SpVIr on the left side was the TeLC-silenced side, and the subnucleus on right side was intact and used as control.

Vibrissa tracking

We tracked the vibrissa curvature across 167 behavioral sessions in which mice contacted the object (77 sessions for the SpVIr-silenced side and 90 sessions for the intact side). Each session contains multiple episodes of object contacts, as well as whisking in air or quiet resting periods. We applied a custom vibrissa tracking algorithm, similar to that in Clack et al. (2012), to measure the curvature of the vibrissae in the successive video frames. Since tracking gives rise to the sequence of points that represent the vibrissa shape, we can directly calculate the curvature within a plane, denoted κ, at any point along the vibrissa. We use

κ=y″(1+y′2)32

where y' = dy/dx. We calculated the averaged curvature of three points along the contact section, and then corrected the calculation based on the vibrissa's intrinsic curvature. The curvature traces of each clip were then smoothed by a Gaussian kernel of appropriate variation. All data was analyzed using routines written in Matlab, Mathworks).

Highlights.

  • Vibrissa contact leads to brainstem mediated feedback signals to facial motoneurons.

  • Intrinsic (protraction) motoneurons receive prompt inhibitory and excitatory feedback.

  • Extrinsic (retraction) motoneurons receive excitatory feedback.

  • These disynaptic reflexes are a substrate for fast, top-down modulation of touch.

Acknowledgments

We thank Conrad Foo for assistance with Matlab coding and Ehud Ahissar and David S. Deutsch for making their figures available. This research was supported by grants from the Canadian Institutes of Health Research (grant MT-5877), the National Institute of Mental Health (MH085499), the National Institute of Neurological Disorders and Stroke (NS058668, NS077986, NS101441 and NS0905905), and the National Science Foundation (EAGER - 2144GA).

Footnotes

AUTHOR CONTRIBUTIONS

M. Deschênes, D.K., and F.W. planned the experiments and wrote the manuscript, M-A.B, M. Demers, M. Deschênes, J.L. and J.T. carried out the experiments and all authors analyzed the data.

The authors declare no competing financial interests.

References

  1. Ashwell KW. The adult mouse facial nerve nucleus: morphology and musculotopic organization. J. Anat. 1982;135:531–538. [PMC free article] [PubMed] [Google Scholar]
  2. Avendaño C, Machín R, Bermejo PE, Lagares A. Neuron numbers in the sensory trigeminal nuclei of the rat: A GABA- and glycine-immunocytochemical and stereological analysis. J. Comp. Neurol. 2005;493:538–553. doi: 10.1002/cne.20778. [DOI] [PubMed] [Google Scholar]
  3. Bae YC, Ihn HJ, Park MJ, Kim HN, Park SK, Bae JY, Lee HW, Kim KH, Yoshida A, Shigenaga Y. The synaptic microcircuitry associated with primary afferent terminals in the interpolaris and caudalis trigeminal sensory complex. Brain Res. 2005;1060:118–125. doi: 10.1016/j.brainres.2005.08.042. [DOI] [PubMed] [Google Scholar]
  4. Bae YC, Ihn HJ, Park MJ, Otterson OP, Moritani M, Yoshida A, Yoshio S. Identification of signal substances in synapses made between primary afferents and their associated axon terminals in the rat trigeminal sensory nuclei. J. Comp. Neurol. 2000;418:299–309. [PubMed] [Google Scholar]
  5. Berg RW, Kleinfeld D. Rhythmic whisking by rat: retraction as well as protraction of the vibrissae is under active muscular control. J. Neurophysiol. 2003;89:104–117. doi: 10.1152/jn.00600.2002. [DOI] [PubMed] [Google Scholar]
  6. Bosman LW, Houweling AR, Owens CB, Tanke N, Shevchouk OT, Rahmati N, Teunissen WH, Ju C, Gong W, Koekkoek SK, De Zeeuw CI. Anatomical pathways involved in generating and sensing rhythmic whisker movements. Front. Integr. Neurosci. 2011;5:1–28. doi: 10.3389/fnint.2011.00053. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Callado-Pérez A, Kleinfeld D, Deschênes M. Subnucleus muralis: first relay for sensory input from the glabrous orofacial skin. Soc. Neurosci. 2016 Abstract 618.04. [Google Scholar]
  8. Chiala NL, Bennett-Clarke CA, Rhoades RW. Differential effects of peripheral damage on vibrissa-related patterns in trigeminal nucleus principalis, subnucleus interpolaris, and subnucleus caudalis. Neurosci. 1992;49:141–156. doi: 10.1016/0306-4522(92)90082-d. [DOI] [PubMed] [Google Scholar]
  9. Clack NG, O'Connor DH, Huber D, Petreanu L, Hires A, Peron S, Svoboda K, Myers EW. Automated tracking of whiskers in videos of head fixed rodents. PLoS Comput. Biol. 2012;8:e1002591. doi: 10.1371/journal.pcbi.1002591. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Deschênes M, Takatoh J, Kurnikova S, Moore JD, Demers M, Elbaz M, Furuta T, Wang F, Kleinfeld D. Inhibition, not excitation, drives rhythmic whisking. Neuron. 2016;90:374–387. doi: 10.1016/j.neuron.2016.03.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Deutsch D, Pietr M, Knutsen PM, Ahissar E, Schneidman E. Fast Feedback in Active Sensing: Touch-induced changes to whisker-object interaction. PLoS ONE. 2012;7:e44272. doi: 10.1371/journal.pone.0044272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Friauf E. Morphology of motoneurons in different subdivisions of the rat facial nucleus stained intracellularly with horseradish peroxidase. J. Comp. Neurol. 1986;253:231–241. doi: 10.1002/cne.902530209. [DOI] [PubMed] [Google Scholar]
  13. Furutani R, Izawa T, Sugita S. Distribution of facial motoneurons innervating the common facial muscles of the rabbit and rat. Okajimas Folia Anat. Jpn. 2004;81:101–108. doi: 10.2535/ofaj.81.101. [DOI] [PubMed] [Google Scholar]
  14. Furuta T, Timofeeva E, Nakamura K, Okamoto-Furuta K, Togo M, Kaneko T, Deschênes M. Inhibitory gating of vibrissal inputs in the brainstem. J. Neurosci. 2008;28:1789–1797. doi: 10.1523/JNEUROSCI.4627-07.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gao P, Hattox AM, Jones LM, Keller A, Zeigler HP. Whisker motor cortex ablation and whisker movement patterns. Somatosens. Mot. Res. 2003;20:191–198. doi: 10.1080/08990220310001622924. [DOI] [PubMed] [Google Scholar]
  16. Ge SN, Li ZH, Tang J, Ma Y, Hioki H, Zhang T, Lu YC, Zhang FX, Mizuno N, Kaneko T, et al. Differential expression of VGLUT1 or VGLUT2 in the trigeminothalamic or trigeminocerebellar projection neurons in the rat. Brain Struct. Funct. 2014;219:211–229. doi: 10.1007/s00429-012-0495-1. [DOI] [PubMed] [Google Scholar]
  17. Gordon G, Fonio E, and Ahissar E. Emergent exploration via novelty management. J. Neurosci. 2014;34:12646–12661. doi: 10.1523/JNEUROSCI.1872-14.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Grant AR, Mitchinson B, Fox CW, Prescott TJ. Active touch sensing in the rat: Anticipatory and regulatory control of whisker movements during surface exploration. J. Neurophysiol. 2008;101:862–874. doi: 10.1152/jn.90783.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Grant RA, Haidarliu S, Kennerley NJ, Prescott TJ. The evolution of active vibrissal sensing in mammals: evidence from vibrissal musculature and function in the marsupial opossum Monodelphis domestica. J. Exp. Biol. 2013;216:3483–3494. doi: 10.1242/jeb.087452. [DOI] [PubMed] [Google Scholar]
  20. Hattox AM, Priest CA, Keller A. Functional circuitry involved in the regulation of whisker movements. J. Comp Neurol. 2002;442:266–276. doi: 10.1002/cne.10089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Herfst LJ, Brecht M. Whisker movements evoked by stimulation of single motor neurons in the facial nucleus of the rat. J. Neurophysiol. 2008;99:2821–2832. doi: 10.1152/jn.01014.2007. [DOI] [PubMed] [Google Scholar]
  22. Hill DN, Bermejo R, Zeigler HP, Kleinfeld D. Biomechanics of the vibrissa motor plant in rat: rhythmic whisking consists of triphasic neuromuscular activity. J. Neurosci. 2008;28:3438–3455. doi: 10.1523/JNEUROSCI.5008-07.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hinrichsen CF, Watson CD. The facial nucleus of the rat: representation of facial muscles revealed by retrograde transport of horseradish peroxidase. Anat. Rec. 1984;209:407–415. doi: 10.1002/ar.1092090321. [DOI] [PubMed] [Google Scholar]
  24. Hires SA, Pammer L, Svoboda K, Golomb D. Tapered whiskers are required for active tactile sensation. eLife. 2013;2:e01350. doi: 10.7554/eLife.01350. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ide LS, Killackey HP. Fine structural survey of the rat’s brainstem sensory trigeminal complex. J. Comp. Neurol. 1985;235:145–168. doi: 10.1002/cne.902350202. [DOI] [PubMed] [Google Scholar]
  26. Klein BG, Rhoades RW. Representation of whisker follicle intrinsic musculature in the facial motor nucleus of the rat. J. Comp. Neurol. 1985;232:55–69. doi: 10.1002/cne.902320106. [DOI] [PubMed] [Google Scholar]
  27. Komiyama M, Shibata H, Susuki T. Somatotopic representation of facial muscles within the facial nucleus of the mouse. A study using the retrograde horseradish peroxidase and cell degeneration techniques. Brain Behav. Evol. 1984;24:144–151. doi: 10.1159/000121312. [DOI] [PubMed] [Google Scholar]
  28. Liang F, Hatanaka Y, Saito H, Yamamori T, Hashikawa T. Differential expression of gamma-aminobutyric acid type B receptor-1a and -1b mRNA variants in GABA and non-GABAergic neurons of the rat brain. J. Comp. Neurol. 2000;416:475–495. [PubMed] [Google Scholar]
  29. Matthews DW, Deschênes M, Furuta T, Moore JD, Wang F, Karten HJ, Kleinfeld D. Feedback in the brainstem: an excitatory disynaptic pathway for control of whisking. J. Comp. Neurol. 2015;523:921–942. doi: 10.1002/cne.23724. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mitchinson B, Martin CJ, Grant RA, Prescott TJ. Feedback control in active sensing: rat exploratory whisking is modulated by environmental contact. Proc. R. Soc. B-Bio. Sci. 2007;274:1035–1041. doi: 10.1098/rspb.2006.0347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Moore JD, Deschênes M, Furuta T, Huber D, Smear MC, Demers M, Kleinfeld D. Hierarchy of orofacial rhythms revealed through whisking and breathing. Nature. 2013;469:53–57. doi: 10.1038/nature12076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Moore JD, Mercer Lindsay N, Deschênes M, Kleinfeld D. Vibrissa self-motion and touch are reliably encoded along the same somatosensory pathway from brainstem through thalamus. PLoS Biol. 2015;13(9):e1002253. doi: 10.1371/journal.pbio.1002253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Nguyen QT, Kleinfeld D. Positive feedback in a brainstem tactile sensorimotor loop. Neuron. 2005;45:447–457. doi: 10.1016/j.neuron.2004.12.042. [DOI] [PubMed] [Google Scholar]
  34. Nguyen QT, Wessel R, Kleinfeld D. Developmental regulation of active and passive membrane properties in rat vibrissa motoneurones. J. Physiol. 2004;556:203–219. doi: 10.1113/jphysiol.2003.060087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pinganaud G, Bernat I, Buisseret P, Buisseret-Delmas C. Trigeminal projections to hypoglossal and facial motor nuclei in the rat. J. Comp. Neurol. 1999;415:91–104. doi: 10.1002/(sici)1096-9861(19991206)415:1<91::aid-cne7>3.0.co;2-a. [DOI] [PubMed] [Google Scholar]
  36. Rossignol S, Dubuc R, Gossard JP. Dynamic sensorimotor interactions in locomotion. Physiol. Rev. 2006;86:89–154. doi: 10.1152/physrev.00028.2005. [DOI] [PubMed] [Google Scholar]
  37. Sachdev RHS, Berg RW, Chompney G, Kleinfeld D, Ebner FF. Unilateral vibrissa contact: Changes in amplitude but not the timing of vibrissa movement. Somat. Mot. Res. 2003;20:162–169. doi: 10.1080/08990220311000405208. [DOI] [PubMed] [Google Scholar]
  38. Sherman D, Oram T, Deutsch D, Gordon G, Ahissar E, Harel S. Tactile modulation of whisking via the brainstem loop: Statechart modeling and experimental validation. PLoS ONE. 2013;8:e79831. doi: 10.1371/journal.pone.0079831. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Sherman D, Oram T, Harel D, Ahissar E. Attention robustly gates a closed-loop touch reflex. Curr. Biol. 2017 doi: 10.1016/j.cub.2017.05.058. epub ahead of print. [DOI] [PubMed] [Google Scholar]
  40. Stanek E, Rodriguez E, Zhao S, Han BX, Wang F. Supratrigeminal bilaterally projecting neurons maintain basal tone and enable bilateral phasic activation of jaw-closing muscles. J. Neurosci. 2016;36:7663–7675. doi: 10.1523/JNEUROSCI.0839-16.2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sreenivasan V, Karmakar K, Rijli F, Petersen CH. Parallel pathways from motor and somatosensory cortex for controlling whisker movements in mice. Eur. J. Neurosci. 2015;11:354–367. doi: 10.1111/ejn.12800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Sherman D, Oram T, Harel D, Ahissar E. Attention robustly gates a closed-loop touch reflex. Curr. Bio. 2017 doi: 10.1016/j.cub.2017.05.058. in press. [DOI] [PubMed] [Google Scholar]
  43. Takatoh J, Nelson A, Zhou X, Bolton MM, Ehlers MD, Arenkiel BR, Mooney R, Wang F. New modules are added to vibrissal premotor circuitry with the emergence of exploratory whisking. Neuron. 2013;77:346–360. doi: 10.1016/j.neuron.2012.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tonomura S, Ebara S, Bagdasarian K, Uta D, Ahissar E, Meir I, Lampl I, Kuroda D, Furuta T, Furue H, Kumamoto K. Structure-function correlations of rat trigeminal primary neurons: Emphasis on club-like endings, a vibrissal mechanoreceptor. Proc. Jasp. Acad. Ser. B Phys. Biol. Sci. 2015;91:560–576. doi: 10.2183/pjab.91.560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Voigts J, Herman DH, Celikel T. Tactile object localization by anticipatory whisker motion. J. Neurophysiol. 2015;113:620–632. doi: 10.1152/jn.00241.2014. [DOI] [PubMed] [Google Scholar]
  46. Wang P, Chen T, Sakurai K, Han BX, He Z, Feng G, Wang F. Intersectional Cre driver lines generated using split-intein mediated split-Cre reconstitution. Sci. Rep. 2012;2:497. doi: 10.1038/srep00497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Watson CR, Sakai S, Armstrong W. Organization of the facial nucleus in the rat. Brain Behav. Evol. 1982;20:19–28. doi: 10.1159/000121578. [DOI] [PubMed] [Google Scholar]
  48. Welker WI. Analysis of sniffing of the albino rat. Behaviour. 1964;12:223–244. [Google Scholar]
  49. Wickersham IR, Finke S, Conzelmann KK, Callaway EM. Retrograde neuronal tracing with a deletion-mutant rabies virus. Nature Meth. 2007;4:47–49. doi: 10.1038/NMETH999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Zhang Y, Zhao S, Rodriguez E, Takatoh J, Han BX, Zhou X, Wang F. Identifying local and descending inputs for primary sensory neurons. J Clin. Invest. 2015;125:3782–3794. doi: 10.1172/JCI81156. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES