Unilateral lingual nerve transection alters jaw-tongue coordination during mastication in pigs

Abstract

During chewing, movements and deformations of the tongue are coordinated with jaw movements to manage and manipulate the bolus and avoid injury. Individuals with injuries to the lingual nerve report both tongue injuries due to biting and difficulties in chewing, primarily because of impaired bolus management, suggesting that jaw-tongue coordination relies on intact lingual afferents. Here, we investigate how unilateral lingual nerve (LN) transection affects jaw-tongue coordination in an animal model (pig, Sus scrofa). Temporal coordination between jaw pitch (opening-closing) and 1) anteroposterior tongue position (i.e., protraction-retraction), 2) anteroposterior tongue length, and 3) mediolateral tongue width was compared between pre- and post-LN transection using cross-correlation analyses. Overall, following LN transection, the lag between jaw pitch and the majority of tongue kinematics decreased significantly, demonstrating that sensory loss from the tongue alters jaw-tongue coordination. In addition, decrease in jaw-tongue lag suggests that, following LN transection, tongue movements and deformations occur earlier in the gape cycle than when the lingual sensory afferents are intact. If the velocity of tongue movements and deformations remains constant, earlier occurrence can reflect less pronounced movements, possibly to avoid injuries. The results of this study demonstrate that lingual afferents participate in chewing by assisting with coordinating the timing of jaw and tongue movements. The observed changes may affect bolus management performance and/or may represent protective strategies because of altered somatosensory awareness of the tongue.

NEW & NOTEWORTHY Chewing requires coordination between tongue and jaw movements. We compared the coordination of tongue movements and deformation relative to jaw opening-closing movements pre- and post-lingual nerve transection during chewing in pigs. These experiments reveal that the timing of jaw-tongue coordination is altered following unilateral disruption of sensory information from the tongue. Therefore, maintenance of jaw-tongue coordination requires bilateral sensory information from the tongue.

INTRODUCTION

During mastication, rhythmic sequential gape cycles provide the kinematic and mechanical framework for bolus processing in the oral cavity. However, the tongue also plays an essential role by supporting the bolus on the teeth as the jaw opens and closes. As food reduction and bolus processing continues, the tongue constantly manipulates and reshapes the bolus against the hard palate and lingual surfaces of the teeth. Despite these complex functions of the tongue occurring on a fairly predictable series of jaw movements, there appears to be a fairly consistent coordination between the opening and closing of the jaw and the movement and deformation of the tongue in humans and other mammals (18, 19, 27, 37, 40). For example, in a typical masticatory gape cycle in pigs, tongue protraction, elongation, and widening occur during the slow close (i.e., power stroke) through slow open phases of the gape cycle, followed by retraction during fast open and fast close phases (18–20).

Tongue-jaw coordination is primarily the result of reflexive control of tongue position by jaw position mediated in the brainstem by direct connections between hypoglossal premotor neurons and trigeminal (CN V) afferents from jaw muscle spindles (11, 21, 22). However, tongue and jaw movements are flexible (i.e., variable in response to food properties) both in timing and amplitude to ensure proper bolus processing and management (10, 24, 30, 37, 46). Although it is well established that information conveyed by oral sensory receptors, such as periodontal mechanoreceptors, alter the kinematics of the jaw during chewing allowing for fine adjustments in the timing or amplitude of movements (41, 42), we know far less about how lingual nerve (LN) afferents modulate or alter tongue movements and deformations in relation to jaw movements. The LN conveys general sensation from the anterior 2/3 of the tongue, serving two populations of lingual mechanoreceptors that are functionally important during feeding: 1) densely populated superficial receptors that provide the tongue with its extreme tactile sensitivity and 2) deep receptors conveying information about tongue position in the absence of tactile stimuli (43). Synaptic coupling between these receptors and ipsilateral jaw-muscle motoneurons ensures precision and safety of tongue movements in coordination with the jaw (23, 44). Moreover, the central nervous system uses information conveyed by the LN to control tongue shape and pressure so that appropriate bolus consistency is reached and proper lingual propulsive forces are generated during the swallow (36). Thus, it is not surprising that the anterior portion of the tongue, which is in constant contact with the food during the oral preparatory phase, is more highly modulated than the glossopharyngeal-innervated posterior portion (28).

There is evidence that the anterior tongue plays a role in modulating tongue and jaw movements and thus jaw-tongue coordination during feeding. In opossums, contact between the tongue and solid foods elicits the jaw-opening reflex, and, in rats, unilateral crush LN injuries cause immediate loss of the jaw-opening reflex (29, 39). In both humans and in animal models, altered lingual sensation impacts cortical motor and sensory centers. For example, following LN transection in rats, the face motor cortex exhibits changes in as little as 1 wk but is more pronounced after 21 days. This may reflect motor compensation strategies resulting from adaptations in motor output in response to altered sensory input (1). In humans, short-term topical anesthesia decreases primary sensory and motor cortex activation, impacting the control of swallowing (38). Finally, in patients with unilateral LN injuries, which are often caused during routine oral procedures (e.g., tooth extractions, injections, and osteotomies), a common complaint is impairment in eating, drinking, and other oral behaviors, and this appears to be associated with altered oromotor performance (e.g., 13, 14, 31). Thus there is evidence that even unilateral disruption of lingual sensory afferents can have a significant clinical impact for behaviors that are bilaterally coordinated.

To date, no studies have systematically evaluated whether impacts on jaw-tongue coordination contribute to reduced oromotor performance following unilateral LN injury. In a previous study, we characterized the effects of LN transection on jaw movements during chewing in an animal model (pigs, Sus scrofa; see Ref. 25). We showed that, across individuals, the amplitude of jaw movements was more adversely affected than their timing. Specifically, we found jaw opening and condylar protraction-retraction increased, whereas jaw yaw (contributing to the transverse component of occlusion) decreased. We interpreted these results to suggest that maintaining the timing but altering the amplitude of jaw movement was a mechanism to maintain coordination with the tongue during chewing (25).

Here, we use a subset of this previously analyzed data set to compare jaw movements before and after unilateral LN denervation with simultaneously recorded data on tongue movements and deformations. This allows us to evaluate whether there are predictable changes in the coordination of the tongue and jaw during chewing. We hypothesize that tongue protraction and retraction will maintain coordination with jaw opening and closing pre- and postinjury to avoid injury to the tongue as the jaw closes. Following LN transection, we expect changes in tongue length and width will be variable with respect to the gape cycle because these deformations likely reflect specific food-handling and manipulation behaviors that require integration of information arising from lingual mechanoreceptors when the bolus touches the tongue’s surface. Finally, we hypothesize that, following the denervation, jaw-tongue coordination will be more disrupted on the side ipsilateral to the transection.

MATERIALS AND METHODS

Data Collection

Jaw movements were quantified in five female pigs (2–3 mo old) using marker-based X-ray Reconstruction of Moving Morphology (XROMM) following our previously described procedures (24, 25). Sterilized 1.0-mm-diameter tantalum markers (Bal-Tec, Los Angeles, CA) were implanted in the cranium and jaw while animals were under general anesthesia (isoflurane 1–5% balance O₂) and with local anesthesia (bupivacaine HCl) injected subcutaneously at each implantation site. Following surgical exposure of the target implant location, a 1.0-mm hole was drilled in the bone. The sterilized markers were gently pushed in the hole, and the incision was closed with either suture or tissue adhesive. Additional beads were placed in the teeth by using a dental drill to produce a hole in the enamel. A minimum of six beads was placed in the jaw and cranium, including tooth beads, to facilitate 3D reconstruction of movement. Animals were given a single intramuscular injection of ketoprofen (2 mg/kg) following surgery.

To characterize tongue movements and deformations, four radiopaque markers (1.0 mm diameter) were implanted in the anterior two-thirds of the tongue to form a rectangle (Fig. 1).The two beads in the anterior part of the tongue were placed approximately one-third of the length of the tongue from the tongue tip when the tongue is relaxed. The two posterior beads were placed ~1 cm anterior to the terminal sulcus. Each bead was placed ~1.0 cm from the lateral edge of the tongue. To implant the beads, the beads were inserted in the tip of a sterile 1-in. 16-gauge needle that was then inserted in the tongue up to the hub. A sterile wire was inserted in the back of the needle to push the bead into the flesh of the tongue. This procedure was done while the tongue was gently pulled out of the mouth so as to not stretch it and thus significantly decrease its dorsoventral height. Care was taken to position the bead to the same depth and configuration in each animal.

Fig. 1.

Marker placement in the tongue. Superior (A) and lateral (B) views of the tongue and jaw of pig 9 showing marker placement. In the superior view, the implanted markers are shown in red, and the virtual markers are shown in yellow. For clarity, only the implanted markers are shown in the lateral view. The Anatomical Coordinate System used to calculate the 3D coordinates of each tongue marker is represented in both views. Markers in the jaw are not shown.

Following all bead implants, recovery, and training, control data were collected using high-speed biplanar fluoroscopy in at least two recording sessions per animal over the course of 1 wk. High-speed biplanar fluoroscopy was recorded at 250 frames/s using two synchronized high-speed video cameras (Oqus 310; Qualisys, Göteborg, Sweden) mounted on the output ports of two fluoroscopes [OEC-9000 (General Electrics, Boston, MA) refurbished by Radiological Imaging Services (Hamburg, PA)]. On average, radiation exposures were set at 80 kVp and 4.5 mA to provide sufficient contrast between markers and bones. Recordings were initiated after chewing commenced. However, over the course of a single feeding sequence, pigs may ingest an additional piece or swallow. In these cases, we attempted to selected for analysis the portion of the sequence consisting of rhythmic chews to the exclusion of these other behaviors. All data are from chewing on size-standardized (2 × 2 × 1 cm) blocks of apple.

In this experimental design, each animal serves as its own control to compare with treatment data, with the treatment being the complete transection of the left LN (branch of the mandibular division of the CN V₃). The left LN was transected in each animal under general anesthesia (isoflurane 1–5% balance O₂) following procedures described previously (25). Briefly, the tongue was gently reflected to the right side to expose the floor of the oral cavity on the left side. An injection of bupivacaine hydrochloride (0.5 mL) was given submucosally for local anesthesia. An incision was made in the mucous membrane running anteroposteriorly at the level of the premolars. The LN was identified and isolated from surrounding tissues. Two suture loops were tied around the LN as proximally as possible to make sure that all branches were included. The nerve was completely transected in between the suture with a scalpel, and the two ends were pushed away from each other. The incision was then sutured closed. After a minimum of 24 h of recovery, treatment data were collected in the same manner as the control data during multiple recording sessions over 1 wk. Following the final data collection session, animals were anesthetized (isoflurane 3–5%) and then euthanized with pentobarbital sodium (1 mL/10 lb iv). According to the standard XROMM workflow, precision thresholds were measured for each animal by imaging the frozen specimens postmortem. Precision and measurement errors for each animal are provided in Supplemental Table S1 (Supplemental data for this article are available at https://doi.org/10.6084/m9.figshare.8242904.v4). All procedures were approved by the Ohio University Institutional Animal Care and Use Committee (protocol no. 12-U-009).

Data Processing

Data processing for jaw movements and quantifying precision and measurement error using XMALab software (version 1.5.0) followed the standard XROMM workflow (2, 24). Distortion inherent to X-ray imaging was corrected by imaging a perforated steel sheet with standardized hole spacing and sizes (part number 9255T641; McMaster-Carr, Robinson, NJ) in each fluoroscopic view. Next, the field of view covered by both fluoroscopes was calibrated before and after each recording session by exposing a custom cube of four plastic sheets containing 64 radiopaque tantalum beads placed in a 4 × 4 fashion 2.5 cm apart from one another.

The screen coordinates of each radiopaque marker (i.e., of the skull, jaw, and tongue) were digitized in each view, and their 3D coordinates were calculated using XMALab software (16). Changes in the coordinates of these markers through time allowed us to reconstruct the 3D position of the skull and the jaw through time, and therefore their respective movements. Skull and jaw models were imported into Maya (Autodesk Inc., San Rafael, CA), and their respective movements were animated frame-by-frame.

A joint coordinate system (JCS) was created to quantify the various degrees of freedom characterizing jaw movements with respect to the skull. Opening and closing jaw movements were extracted as rotation about a transverse axis (z-axis, Rz) running through the temporomandibular joint. Note that, in the context of the present study, only jaw pitch Rz (i.e., the rotation of the jaw around an axis passing through the left and right mandibular condyles) was used because it represents the cyclical opening-closing motion of the jaw as the animal chews. Tongue movements and deformations associated with each gape cycle were visualized by importing the 3D coordinates of the tongue markers as locators in the Maya skeletal animation for each sequence. The 3D coordinates of the tongue markers were parented to the joint coordinate system used to quantify the jaw movements so that tongue movements and deformation could be quantified independent of skull position and motion. Based on the 3D coordinates of these four implanted markers, the length of the left and right side of the tongue was measured as the 3D distance between the anterior and posterior markers on the left and right markers, respectively. Similarly, the width of the anterior and posterior regions of the tongue was measured as the 3D distance between the two anterior and posterior markers, respectively.

In addition, two virtual points were added in the animation: one midway between the anterior markers, the other midway between the posterior markers (see Fig. 1). Because the x-axis was set as the rostrocaudal axis, the x-coordinates of the anterior and posterior virtual markers quantify protraction-retraction of the tongue. An increase in the x-coordinate indicates tongue protraction, whereas a decrease in the x-coordinate indicates tongue retraction. In summary, six kinematic variables were quantified to evaluate tongue movements and deformations: 1) protraction-retraction of the anterior region of the tongue, 2) protraction-retraction of the posterior region of the tongue, 3) tongue length on the left side of the tongue, 4), tongue length on the right side of the tongue, 5) width of the anterior region of the tongue, and 6) width of the posterior region of the tongue. Note that the data set of individual 6 is limited because only the anterior markers could be digitized properly throughout the sequences recorded. As a result, only the data associated with the movements and deformations of the anterior part of the tongue (i.e., protraction-retraction and width) were available for this particular individual.

Data Analysis

To investigate jaw-tongue coordination in time pre- and post-transection, we conducted cross-correlation analysis of the kinematic waves representing tongue movements and deformations with respect to the one kinematic wave representing jaw opening-closing motion (i.e., jaw pitch, Rz) used as the reference wave. Cross-correlation is an established method for investigating coordination of movements represented by time series (17, 26). All cross-correlations were conducted using the crosscorr function in Matlab (MathWorks, Natick, MA). Cross-correlation analysis calculates the best-fitted lag in time between two kinematic waves. To satisfy the restrictions of cross-correlation analyses, only sequences of 300 or more frames of chewing were included in the present study. This corresponds to sequences of four or more consecutive chewing cycles. A time lag of zero indicates that the waves are synchronous, a negative time lag indicates that the kinematic wave analyzed precedes the reference wave, and finally a positive time lag indicates that the kinematic wave analyzed is delayed with respect to the reference wave.

To test the effects of LN transection on jaw-tongue coordination during chewing, the time lags between the jaw pitch wave and each of the six tongue waves were compared before (i.e., control) and after transection using a MANOVA associated with univariate F ratios with the following design: treatment as a fixed factor, individuals as a random factor, and the associated treatment × individual interaction term. A significant treatment × individual interaction term indicates that individuals were not affected by LN transection in a similar manner. In those cases, treatment effects were tested at the individual level (i.e., for each individual separately). In this analysis, the critical significance level ɑ was set at 0.05, but, because six time lags were tested overall, the significance threshold of the P value was adjusted to 0.0083 following Bonferroni’s correction (P = ɑ/n where n is the number of tests; see Ref. 3). All statistical analyses were performed using custom-built scripts in R Studio (Boston, MA).

RESULTS

General Description of Tongue Movements and Deformations Relative to the Gape Cycle Pre- and Post-LN Transection

Representative kinematic waves illustrating anteroposterior tongue position and deformations before and after LN transection in the same animal are provided in Figs. 2 and 3, respectively, over a 1.2-s window comprised of four sequential chewing cycles. With the LN intact, at maximum gape, the tongue is in the process of retracting (Fig. 2A) and shortening (Fig. 2B). As the jaw closes, the tongue continues to retract and may marginally shorten further. Both maximum tongue retraction and maximum shortening occur during jaw closing, usually midway through slow close and almost always just before minimum gape. At the end of closing and throughout fast open, the tongue protracts and elongates until it reaches its maximum length and protraction near the transition between slow and fast open. Finally, during late opening, the tongue quickly retracts and shortens. For both anteroposterior tongue position and deformation, the anterior and posterior regions of the tongue change more or less at the same time, that is, the anterior and posterior regions protract and retract together, and the left and right sides of the tongue are lengthening and shortening together.

Fig. 2.

Representative kinematic profiles of tongue movements and deformations relative to jaw pitch (Rz) in animal 13 before lingual nerve transection (i.e., control data). Anteroposterior protraction-retraction of the anterior and posterior regions of the tongue (orange and red, respectively, A), left and right length (light and dark green, respectively, B), and width of the anterior and posterior regions of the tongue (light and dark blue, respectively, C). In A, increasing values represent tongue protraction, and decreasing values represent tongue retraction. In each graph, an increase in jaw pitch (Rz, in black) represents jaw closing, and a decrease in jaw pitch represents jaw opening. In accordance with changes in jaw pitch, the phases of the chewing cycles are shown for a selected cycle (FC, fast closing; SC, slow closing; SO, slow opening, and FO, fast opening).

Fig. 3.

Representative kinematic profiles of tongue movements and deformations relative to jaw pitch (Rz) in animal 13 after lingual nerve transection (i.e., treatment data). Anteroposterior protraction-retraction of the anterior and posterior regions of the tongue (orange and red, respectively, A), left and right length (light and dark green, respectively, B), and width of the anterior and posterior regions of the tongue (sky and dark blue, respectively, C). In A, increasing values represent tongue protraction, and decreasing values represent tongue retraction. In each graph, an increase in jaw pitch (Rz, in black) represents jaw closing, and a decrease in jaw pitch represents jaw opening. In accordance to changes in jaw pitch, the phases of the chewing cycles are shown for a selected cycle (FC, fast closing; SC, slow closing; SO, slow opening, and FO, fast opening).

In terms of width, however, the anterior and posterior regions of the tongue show a less coordinated pattern both to each other and with respect to jaw pitch (Fig. 2C). The anterior region of the tongue deforms little during early closing and then starts to narrow during slow close, although in some cycles it widens slightly (Fig. 2C). It typically reaches its narrowest width during slow open and then widens just before the transition between slow and fast open. Maximum width of the anterior tongue usually plateaus around maximum gape. Note that, in some gape cycles, however, this plateau may include a rapid narrowing of the tongue around the time of maximum gape. In contrast, the posterior region of the tongue usually narrows during early jaw closing until it reaches minimum width around the fast close-slow close transition or later during slow close until the time of minimum gape. At the end of slow close and into slow open, the posterior tongue widens. Accordingly, in most cycles, maximum width of the posterior region of the tongue occurs during slow open about the time of minimum width of the anterior tongue. Subsequently, during fast open, the posterior tongue begins to narrow.

Following LN transection, protraction-retraction dynamics of the anterior and posterior regions of the tongue were altered such that the two regions did not always track each other as they did pretransection, and their coordination with jaw movements was less consistent cycle-to-cycle (Fig. 3A). Some cycles are characterized by very little change in anteroposterior tongue position, whereas some others are characterized by substantial changes. Changes in tongue length and, to a lesser extent, tongue width, relative to jaw movements, on the other hand, were generally similar between control and LN transection sequences. Moreover, for tongue length, the left and right sides of the tongue are lengthening and shortening together (Fig. 3B). For tongue width, we again observed that the anterior and posterior regions of the tongue exhibit almost opposite deformation patterns (Fig. 3C). Nevertheless, quantitative analysis (see below) of the timing of tongue movements and deformations relative to jaw pitch revealed some interesting differences between control sequences and those following unilateral LN transection.

Effect of Unilateral LN Transection on the Coordination Between Tongue Movements and Deformations and Jaw Pitch

Protraction-retraction.

When the LN is intact, the maximally correlated cross-correlation function (CCF) indicates that protraction-retraction of both the anterior and posterior part of the tongue occurs after changes in jaw pitch (Table 1 and Fig. 4). For the anterior tongue, the lag in milliseconds of the maximum CCF shifts from 53.8 (SD 49.2) before the transection to −54.5 (SD 47.8) after the LN transection. This difference in time lag is significantly different between treatments [F(1,37) = 56.74, P < 0.001], but is not affected by a significant treatment × individual or by significant differences between individuals. For the posterior tongue, the lag between jaw protraction-retraction and jaw pitch averages 67.3 ms (SD 37.3) when the LN is intact and −56.7 (SD 53.7) after the LN transection (Table 1 and Fig. 4). This difference in lag is significant between treatments [F(1,31) = 95.43, P < 0.001], with no treatment × individual interaction, but individual differences were significant [F(3,31) = 5.96, P = 0.015].

Table 1.

Summary of the lags (in ms) between jaw pitch and tongue movements and deformations pre- and post-LN transection

	Control (Nsequence = 20) (Ncycle = 263)		Transection (Nsequence = 22) (Ncycle = 278)		Treatment Effect	Individual Differences
	Lag (SD)	r (SD)	Lag (SD)	r (SD)
Protraction-retraction
Anterior	53.8 (49.2)	0.488 (0.245)	−54.5 (47.8)	0.468 (0.241)	F(1,37) = 56.74, P < 0.001	NS
Posterior	67.3 (37.3)	0.512 (0.241)	−56.7 (53.7)	0.481 (0.234)	F(1,31) = 95.43, P < 0.001	F(3,31) = 5.96, P = 0.015
Tongue length
Left	46.4 (38.7)	0.521 (0.194)	−39.6 (53.1)	0.443 (0.242)	F(1,31) = 49.19, P = 0.002	F(3,31) = 10.77, P = 0.01
Right	45.7 (37.2)	0.502 (0.207)	29.1 (25.8)	0.569 (0.176)	NS	F(3,31) = 7.07, P = 0.01
Tongue width
Anterior	45.6 (52.4)	0.495 (0.257)	−83.8 (44.4)	0.516 (0.230)	F(1,37) = 82.63, P < 0.001	NS
Posterior	68.0 (61.9)	0.527 (0.236)	−75.3 (70.6)	0.508 (0.200)	Individuals 5, 10, and 13

Values are means and standard deviations (SD) of the highest significantly correlated lags across sequences for all five individuals combined. Positive time lags indicate that tongue movement is delayed relative to jaw pitch; negative lags indicate that tongue movement precedes jaw pitch. N_sequence is the total number of sequences analyzed. N_cycle is the total number of chewing cycles included in the analysis. r, correlation coefficient; NS, nonsignificant. Sequence-level data and individual observations are provided in Supplemental Table S2 (https://doi.org/10.6084/m9.figshare.8242904.v4).

Fig. 4.

Effects of unilateral lingual nerve transection on the protraction-retraction of the anterior (A) and posterior (B) region of the tongue in pigs. Boxplots and sequence-level lags (circles) for control (gray; n = 20) with treatment data (white; n = 22) pooled (all) and for each individual. The protraction-retraction movements of the tongue occurred significantly earlier with respect to jaw pitch after the transection of the lingual nerve, indicating that tongue-jaw coordination is significantly altered after the loss of lingual afferents. *Significant differences between control and treatment.

Tongue length.

The maximum CCF pre- and post-transection revealed differences in the coordination of the left and right sides of the tongue with changes in jaw pitch. Changes in left anteroposterior tongue deformation (i.e., in length) are delayed 46.4 ms (SD = 38.7) relative to changes in jaw pitch in controls, whereas after the left LN transection these deformations precede changes in jaw pitch by 39.6 ms (SD = 53.1). This difference in lag is significantly different between treatments [F(1,31) = 49.19, P = 0.002] and between individuals [F(3,31) = 10.77, P = 0.01], but note that no significant treatment × individual interaction was detected. On the right side of the tongue (i.e., contralateral to the LN transection), changes in tongue length occur ~45.7 ms (SD = 37.2) and 29.1 ms (SD = 25.8) before and after LN transection. This difference in lag is not significantly different between treatments [F(1,31) = 1.82, P = 0.19]. Note that this variable was not affected by a significant treatment × individual interaction but that differences between individuals were significant [F(3,31) = 7.07, P = 0.01]. In other words, unilateral LN transection only altered the coordination of changes in tongue length on the ipsilateral (left) side.

Tongue width.

Before LN transection, changes in anterior and posterior tongue width followed changes in jaw pitch by 45.6 ms (SD = 52.4) and by 68.0 ms (SD = 61.9), respectively. The transection had the same effect on the width of both the anterior and posterior regions, which was to shift deformation before jaw pitch (Table 1 and Fig. 5): −83.8 ms (SD = 44.4) for the anterior tongue and −75.3 ms (SD = 70.6) for the posterior tongue. Whereas the difference in time lag for the anterior tongue is significantly different between treatments [F(1,37) = 82.63, P < 0.001] without significant interaction effects or differences between individuals, the difference in time lag for the posterior tongue has a significant treatment × individual interaction (P = 0.014), with significant treatment effects detected in three of the four individuals [individual 5: F(1,6) = 10.66, P = 0.017; individual 10: F(1,8) = 17.3, P = 0.003; individual 13: F(1,7) = 73.72, P < 0.001]. In summary, LN transection results in tongue width changes preceding changes in jaw pitch. This is reflected in the change in the sign of the lag from positive to negative, with the difference between pre- and post-transection lags averaging >100 ms (Table 1 and Fig. 6).

Fig. 5.

Effects of unilateral lingual nerve transection on the changes in length of the left (i.e., ipsilateral; A) and right (i.e., contralateral; B) side of the tongue in pigs. Boxplots and sequence-level lags (circles) for control (gray; n = 20) with treatment data (white; n = 22) pooled (all) and for each individual. Differences pre- and post-transection in the anteroposterior deformation of the tongue (i.e., length) were significant on the ipsilateral side to the transection (i.e., left) but not on the contralateral side (i.e., right). *Significant differences between control and treatment.

Fig. 6.

Effects of unilateral lingual nerve (LN) transection on the width of the anterior (A) and posterior (B) region of the tongue in pigs. Boxplots and sequence-level lags (circles) for control (gray; n = 20) with treatment data (white; n = 22) pooled (all) and for each individual. LN transection significantly altered the coordination between jaw pitch and the width of the anterior tongue. For the posterior tongue, the coordination was only altered in 3 of the 4 individuals (5, 9, 13). *Significant differences between control and treatment.

DISCUSSION

In accordance with previous studies of mammalian chewing (e.g., 18, 19, 27), our control data demonstrate strong coordination between tongue protraction-retraction and jaw opening-closing during chewing in pigs. For example, although tongue retraction has started when the jaw is at maximum gape, further retraction often occurs through jaw closure. The beginning of tongue protraction slightly precedes the initiation of jaw opening, and maximum tongue protraction occurs midway through jaw opening. In humans, the tongue reaches maximum retraction during jaw closing, and maximum protraction occurs during jaw opening (27). Similar patterns have been demonstrated for macaques (8) and hyraces (4). We also observed coordination between anterior tongue deformation and jaw pitch that is partially reminiscent of previous observations (18), albeit with some notable differences. In both studies, anteroposterior deformation of the tongue was coordinated with jaw movements and with protraction-retraction. Likewise, both studies showed that the anterior tongue widens during jaw opening but that there is less of a coordination between changes in tongue width and jaw movements. Nevertheless, our results did not exactly mirror previous observations (18). For example, they found that posterior tongue width increased during jaw opening through the early part of slow close. In our animals, the posterior tongue widens at the end slow close and into slow open. These differences may be related to differences in placement of the posterior beads in the two studies. Whereas our posterior beads were implanted in the tongue approximately midway through the anterior two-thirds of the tongue, the sonomicrometry crystals used the previous study (18) were placed significantly more posteriorly closer to the circumvallate papillae, which are located at the base of the tongue. The combination of these studies suggests that changes in tongue width might occur from rostral to caudal sequentially and/or in a compartmentalized manner during the gape cycle. Work on humans demonstrates that, despite high levels of interdigitation of intrinsic tongue muscles, select regions of these muscles are activated as functional units during vocalization and speech to deform the tongue (33, 47). This may also be the case for the dynamic and nonsymmetrical movements that occur during chewing and other feeding behaviors, which are particularly challenging to characterize.

We initially hypothesized that coordination between tongue protraction-retraction and jaw opening-closing would be maintained following LN transection to avoid self-injury, whereas the coordination of tongue deformations (i.e., length, width) and jaw opening-closing would be altered by LN transection because of unilateral loss of sensory afferents impacting bolus manipulation and handling. Our results show that treatment alters each dimension of jaw-tongue coordination, although the magnitude of the effect varied depending on the kinematic parameter and region of the tongue. Overall, LN transection decreased the lag between jaw pitch and tongue protraction-retraction (see Fig. 4) as well as the lag between jaw pitch and the intrinsic deformations of the tongue, such as its length (see Fig. 5) and its width (see Fig. 6). In fact, the lag between jaw pitch and tongue movements and deformations decreases as much as its sign switched, so that changes in tongue movement and shape shift from following changes in jaw pitch to preceding them after LN transection. In other words, sensory loss from the anterior two-thirds of the tongue forced tongue “cycles” to be completed earlier than jaw pitch cycles, which may indicate that tongue movements and deformations might be incomplete. Additional analysis of the effects of sensory loss on the magnitude of tongue movements and deformations relative to their timing will provide further insights into the effects of tongue injuries on chewing behavior.

As expected, there were significant impacts on the anteroposterior deformation of the tongue on the side ipsilateral to the transection, whereas it had no effect on deformation of the contralateral side. Although pigs bilaterally occlude during chewing, like other mammals, there is a dominant chewing side associated with the direction of jaw movement (7). Our results suggest that intraoral bolus manipulation may be significantly adversely impaired by the loss of lingual afferents during chewing cycles on the ipsilateral side. This could be accommodated for by chewing more on the contralateral side, but this was not consistently observed in our data set (Supplemental Table S1). Moreover, in all of the animals, we observed injuries along the side of the tongue ipsilateral to the LN transection from biting their tongue. Thus, although the tip of the tongue may be protected from injury by altering the anteroposterior movement and mediolateral deformation relative to the jaw, this is not the case for the sides of the tongue. This suggests that compensatory strategies may only partially overcome the challenges of chewing with peripheral disruption to unilateral lingual afferents.

Consistent with previous observations on pigs (6, 24), our animals showed a tendency to alternate chewing sides frequently, oftentimes from one chew to the next within a sequence. Animals did not exhibit a preference for one side over the other, and all analyzed sequences included left and right chews in approximately equal proportions across the sequences, but not necessarily within each sequence (see Supplemental Tables S1 and S2). However, because cross-correlation analyses require continuous waveforms encompassing a sequential series of multiple chewing cycles, we could not consider the effects of chewing side directly. If unilateral LN denervation primarily affects ipsilateral gape cycles and tongue movements, this would be apparent at the level of the chewing cycle and not at the level of the chewing sequence as presented here. The fact that the animals did not respond with longer sequences of chewing on the same side (either ipsilateral or contralateral to the lesion) makes it difficult to fully assess the interaction between chewing side, LN transection, and jaw-tongue modulation. Nevertheless, in a subset of our data from which three to five consecutive same-side chews could be selected within a sequence, we found that the lag between the length changes of the ipsilateral side of the tongue and jaw pitch was 44 ms smaller when chewing ipsilateral vs. contralateral to the lesion. In contrast, length changes on the contralateral side of the tongue showed only a 28-ms reduction in the lag between left and right chews after the transection. Although it is not clear whether this is a functionally significant difference, it does suggest that examination of a larger data set where chewing side is accounted for would provide insight into how chewing side impacts modulation of the tongue and jaw. Further work comparing the effects of unilateral LN transection on ipsilateral and contralateral chews will allow us to explore whether and, if so, how each side of the tongue independently responds to sensory loss.

The results for tongue protraction-retraction and anteroposterior deformation are interesting in light of our previous work on the effects of unilateral lingual deafferentation on jaw movements (25). We previously reported that unilateral LN transection in these same animals altered the magnitude of jaw movements more than their timing during chewing. We interpreted this to mean that maintaining tongue-jaw coordination during feeding is critical. The results from the present study support this interpretation. Furthermore, because the lag between jaw pitch and the width of both the anterior and posterior regions of the tongue was sufficiently pronounced following the transection to reverse the offset for both regions, lingual afferents may be essential for maintaining anteroposterior functional compartmentalization of the tongue relative to jaw movements. This substantial change suggests reduced or altered bolus handling performance both anteriorly and posteriorly.

In conclusion, we have shown that lingual afferents participate in the production of chewing movements by conveying sensory information necessary for coordinating the tongue and the jaw. Moreover, normal coordination of these structures requires bilateral afferent information from the tongue. This study also demonstrates that other oral afferents, such as periodontal, labial, or mucosal mechanoreceptors conveying information about bolus properties or proprioceptors conveying information about tongue positioning within the oral cavity (15, 41), do not fully compensate for the loss of information conveyed by lingual mechanoreceptors to maintain jaw-tongue coordination. Thus, even though sensory cues are collected from a variety of sources within the oral activity, jaw-tongue coordination is disrupted by the unilateral loss of only one of these sources. This conclusion is bolstered by the fact that, in humans, temporarily blocking unilateral LN afferents delays corticomotor control of the tongue (5), and bilateral blocks of LN afferents significantly alter somatosensory awareness (9). Altered somatosensory awareness following local anesthesia has been demonstrated for other oral structures innervated by the trigeminal, including the lips and teeth (45). Although our data cannot address how the LN transection alters underlying sensorimotor integration and ultimately cortical and subcortical control of tongue movements, altered somatosensory awareness of the tongue following the transection in our study may account for the observed differences between treatment and controls. This is because the transection impairs both the rapidly adapting superficial lingual mechanoreceptors that respond to mechanical stimuli from the bolus and surrounding tissues that ultimately provide a sense of “tongue awareness” as well as the slow-adapting deep mechanoreceptors that respond in the absence of tactile stimulation when the tongue is moving relative to surrounding tissues (43).

Finally, some of the observed changes in movements and deformation of the tongue relative to the jaw may be protective or even compensatory in nature to adjust for bolus mishandling. Given the high plasticity in the corticomotor control of tongue muscles (34, 35), teasing apart the basis of the observed changes in tongue movement and deformation following injury is important for understanding oral adaption and pathways for rehabilitation and recovery. Alternatively, in light of evidence from human studies that there are peripheral connections between the lingual and hypoglossal nerves (12, 32, 48), the changes in coordination of tongue movements and deformations as observed here in pigs might also be associated with motor deficits resulting from LN denervation. Although these connections still need to be explored in detail for pigs, given the pronounced changes in tongue movements and motor control that occur with unilateral hypoglossal nerve palsies, these connections to the lingual nerve likely have a minor impact on tongue motor control.

GRANTS

This research was supported by the National Institute of Dental and Craniofacial Research (1R15-DE-023668-01A1) and the National Science Foundation (DBI-0922988 and IOS-1456810).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.J.M., R.A.O., H.C., and S.H.W. performed experiments; S.J.M. and S.H.W. analyzed data; S.J.M. and S.H.W. interpreted results of experiments; S.J.M., R.A.O., and S.H.W. prepared figures; S.J.M. and S.H.W. drafted manuscript; S.J.M., R.A.O., and S.H.W. edited and revised manuscript; S.J.M., R.A.O., H.C., and S.H.W. approved final version of manuscript; S.H.W. conceived and designed research.