Precise movements of the tongue are necessary for licking, suckling, chewing, and swallowing (Lowe, 1981). The intrinsic muscles (contained within the tongue) and most of the extrinsic muscles (attach to the body of the tongue) of the tongue are innervated by motoneuron axons projecting out of the hypoglossal nucleus. The hypoglossal nucleus is divided into two major subcompartments. Motoneuron axons coming out of the dorsal compartment compose the lateral branch of the hypoglossal nerve while the ventral compartment motoneuron axons comprise the medial branch (Krammer et al., 1979; Uemura-Sumi et al., 1988; O'Reilly and FitzGerald, 1990; Sokoloff, 1993; Altschuler et al., 1994; Aldes, 1995). The ventral (protrusor) compartment is composed of the motoneurons that innervate the extrinsic protrusor muscle genioglossus and intrinsic (verticalis and transversus) muscles (Aldes, 1995). The dorsal compartment contains motoneurons that innervate the extrinsic retrusor muscles hyoglossus and styloglossus and other intrinsic (superior longitudinal and inferior longitudinal) muscles (McClung and Goldberg, 1999). In this article, motoneurons innervating the hyoglossus and styloglossus muscles will be referred to as hyoglossus and styloglossus motoneurons.
Previous work from this laboratory (Guo et al., 1996; McClung and Goldberg, 1999) confirmed a bimodal distribution of hyoglossus and styloglossus motoneurons within the dorsal compartment of the hypoglossal nucleus. Specifically, we noted that hyoglossus motoneurons were found from 1.0 mm caudal to 0.4 mm rostral to the internal obex (defined as the junction of the fourth ventricle and the central canal in the transverse plane), whereas styloglossus motoneurons were located 0.5–1.0 mm rostral to the internal obex in the adult rat.
Differences in the direction of the dendritic projections of hyoglossus and styloglossus motoneurons have been reported previously in the rat (Altschuler et al., 1994). These authors found both hyoglossus and styloglossus motoneurons to have their primary dendrites project into the ventral-medial and dorsal-lateral directions. In addition, they found hyoglossus motoneurons to have a more pronounced dendritic projection in the ventral-medial direction. However, they also found an intermingling of hyoglossus and styloglossus motoneurons in both the rostral and caudal halves of the hypoglossal nucleus, which is inconsistent with more recent studies showing separate locations for hyoglossus and styloglossus motoneurons (Guo et al., 1996; McClung and Goldberg, 1999).
Postnatal development of hypoglossal motoneurons has been studied in genioglossus (tongue protrusor) motoneurons only (Brozanski et al., 1989; Cameron et al., 1989; Núñez-Abades et al., 1994; Núñez-Abades and Cameron, 1995), while development of tongue retrusor motoneuron morphology has not been investigated. The postnatal development of genioglossus motoneurons has been used to represent postnatal development of all hypoglossal motoneurons (Berger et al., 1996). Genioglossus motoneurons are in the same relative location within the hypoglossal nucleus in the neonatal and adult rat, suggesting that any migration of motoneurons has already taken place by the first week of life (Sokoloff, 1993). The purpose of this study was to investigate the postnatal development of the hyoglossus and styloglossus (tongue retrusor) motoneurons using 1-, 2-, 3-, and 8-week-old rats. The postnatal development of tongue retrusor motoneurons is of interest because both protrusor and retrusor muscles' roles in swallowing (Miller, 1986). The current findings further our understanding of normal pattern of postnatal development of XII motoneurons and provide essential baseline data for comparison with other groups, i.e., artificially raised rats deprived of suckling activity.
MATERIALS AND METHODS
Virginia Commonwealth University's Institutional Animal Care and Use Committee approved all procedures and protocols for animal care. All animals were housed on a 12-hr light/dark cycle and kept with their mothers until surgery. Following recovery from surgery, the animals were put back with their mothers until perfusion. Those animals that were capable of eating rat chow had unlimited access to standard rat chow and water.
Experiments were carried out on 20 Sprague-Dawley rats consisting of five animals from each of four postnatal age groups: 6–7 days (P6–7), 13–14 days (P13–14), 20–21 days (P20–21), and 49–54 days (P49–54). Age was determined by the day of neural tracer injection (counting the animal's day of birth as postnatal day 1). A surgical plane of anesthesia was created using a combination of ketamine (45 mg/kg, i.p.) and xylazine (6 mg/kg, i.p.) for surgical procedures. Animals were given an overdose of ketamine and xylazine for fixation procedures. Absence of flexor withdrawal was used to assess and maintain appropriate levels of anesthesia.
Using a ventral surgical approach through the midline of the neck, the styloglossus muscle was exposed and 0.4–5.0 μl of 0.1% cholera toxin (subunit B) conjugate of horseradish peroxidase (CTHRP) was injected into the medial portion of the innervation zone. The innervation zone for styloglossus is adjacent to the hyoid bone, where the hyoglossus muscle inserts and no other hypoglossal innervated muscle is in the vicinity. This anatomical configuration allowed us to purposefully inject into styloglossus while CTHRP also spread into the hyoglossus muscle only. Pilot testing had shown that reducing the volume of CTHRP injected into styloglossus would label styloglossus motoneurons only, particularly in the older animals. This isolation of injected CTHRP becomes increasingly more difficult with the younger animals. Therefore, using this new technique, we were able to isolate hyoglossus and styloglossus from intrinsic motoneurons by injecting a small volume of CTHRP into styloglossus. In previous studies, we surgically isolated the motoneurons of interest by injecting CTHRP in the appropriate muscle and cutting the axons to all other muscles in the vicinity. This minimized the possibility of CTHRP from being retrogradely transported back to the hypoglossal nucleus by motoneurons innervating the adjacent musculature. Previous research and our pilot testing have shown a consistent gap between the location of the most rostral hyoglossus motoneuron and the most caudal styloglossus motoneuron (Guo et al., 1996; McClung and Goldberg, 1999). This anatomical phenomenon was used to confirm the selective labeling of hyoglossus and styloglossus motoneurons.
Following a 24-hr survival period for retrograde transport of CTHRP to the hypoglossal nucleus, each animal was deeply anesthetized as previously indicated and killed by transcardial perfusion with a mixed aldehyde fixative. Serial (50 μm) transverse sections of the hypoglossal nucleus were made and CTHRP-labeled motoneurons were histochemically localized with tetramethyl benzidine. Labeled sections were mounted, counterstained in neutral red, coverslipped with Permount, and analyzed under bright-field illumination. A computer-assisted image analysis system was used to outline the cell body and base of the primary dendrites and perform morphometric analyses.
Cross-sectional area, form factor, and direction of dendritic projection were used to assess postnatal development in addition to the location of the motoneuron populations within the hypoglossal nucleus. Form factor refers to how closely the traced cell resembles a perfect circle (a maximum value of 1.0 indicates a perfect circle).
Ninety-five cells from each motoneuron group (hyoglossus and styloglossus) were randomly selected for statistical analysis. A factorial analysis of variance (age × motoneuron group) was used to assess differences between age groups and motoneuron groups. Scheffé's multiple comparisons were used to assess where the differences occur. Level of significance was set to 0.05.
Labeled hyoglossus motoneurons were found in the caudal half of the dorsal subcompartment beside the central canal while styloglossus motoneurons were found in the rostral half of the dorsal subcompartment of the hypoglossal nucleus deep to the fourth ventricle for each age group. Figure 1 shows the lateral location of hyoglossus and styloglossus motoneurons for P6–7 and P49–54 age groups. This was consistent for all age groups. Table 1 displays the cell measurements for hyoglossus and styloglossus motoneurons of each age group. There was a significant interaction between age and motoneuron group for cross-sectional area (F = 3.144; P = 0.025) and form factor (F = 3.616; P = 0.013).
Comparison of Hyoglossus and Styloglossus Motoneurons
No differences were found in cross-sectional area between hyoglossus and their age-matched styloglossus counterparts (Fig. 2). However, the rate of growth during first 2 weeks of life is different. Cross-sectional area of hyoglossus motoneurons during the first week of life was 81% of the area of 8-week-old hyoglossus motoneurons. One-week-old styloglossus motoneurons were 69% of the area for their 8-week-old counterparts. By the end of the second week, hyoglossus and styloglossus motoneurons were 95% and 93% of the area for their 8-week-old counterparts, respectively. As mentioned previously, form factor refers to how closely the contour of the cell resembles a perfect circle. As the animal ages, form factor measures for both hyoglossus and styloglossus motoneurons decline (Fig. 3; F = 15.238; P = 0.003). However, the decline appears more pronounced for the styloglossus motoneurons dropping from 0.58 to 0.45 over the 8-week period. In contrast, the form factor for hyoglossus motoneurons drops from 0.62 to 0.56 within the first 2 weeks, then stabilizes at ∼ 0.52–0.54. Our observations of the primary dendritic pattern suggest that both hyoglossus and styloglossus motoneurons possess some differences in the spatial distribution of their primary dendrites (Fig. 4). Hyoglossus and styloglossus motoneurons project in the dorsal-lateral and ventral-medial directions. Styloglossus motoneurons appear to have fewer dendrites projecting into the ventral-medial direction than their hyoglossus motoneuron counterparts. This pattern of dendritic projection appears to be consistent throughout all age groups.
Table 2 describes the location of labeled hyoglossus motoneurons relative to the obex within the hypoglossal nucleus for each age group. In terms of cross-sectional area, hyoglossus motoneurons reach their adult size by the third week of life (F = 13.667; P = 0.005).
Table 2. Average location of labeled hyoglossus motoneurons relative to the obex within the hypoglossal nucleus
∼ 50–400 μm caudal to obex
∼ 50–400 μm caudal to obex
∼ 350 μm caudal to obex to ∼ 100 μm rostral
∼ 550 μm caudal to obex to ∼ 250 μm rostral
Table 3 describes the location of labeled styloglossus motoneurons relative to the obex within the hypoglossal nucleus for each age group. Styloglossus motoneurons reach their adult size by the second week of life (F = 13.667; P < 0.001).
Table 3. Average location of labeled styloglossus motoneurons relative to the obex within the hypoglossal nucleus
300–500 μm rostral to obex
∼ 300–500 μm rostral to obex
∼ 450–600 μm rostral to obex
∼ 550–850 μm rostral to obex
The purpose of this study was to describe the location and some morphological changes to hyoglossus and styloglossus motoneurons within the dorsal (retrusor) subcompartment of the hypoglossal nucleus during normal postnatal development in the rat. To date, similar research has been performed only on genioglossus motoneurons in the ventral (protrusor) subcompartment of the hypoglossal nucleus in the rat (Brozanski et al., 1989; Mazza et al., 1992; Sokoloff, 1993; Núñez-Abades et al., 1994; Núñez-Abades and Cameron, 1995) and kitten (Cameron et al., 1989).
The first major finding in this study was the segregation of hyoglossus and styloglossus motoneurons within the dorsal (retrusor) subcompartment of the hypoglossal nucleus across all age groups. The hyoglossus and styloglossus motoneurons were located laterally within the hypoglossal nucleus on the transverse sections. In addition, there was a consistent gap between the most rostral hyoglossus motoneurons found and the most caudal styloglossus motoneurons found. This suggests that we were successful in labeling only the hyoglossus and styloglossus motoneurons and not the intrinsic motoneurons located in the dorsal subcompartment of the hypoglossal nucleus (McClung and Goldberg, 1999). This finding is consistent with previous research in our laboratory (Guo et al., 1996; McClung and Goldberg, 1999). Earlier studies have shown coexistence of hyoglossus and styloglossus motoneurons within the rostral and caudal halves of the dorsal subcompartment of the hypoglossal nucleus in the adult rat (Altschuler et al., 1994) and monkey (Sokoloff and Deacon, 1992). We have assumed that this can be explained by spreading of neural tracer into the adjacent musculature. It is through spreading that we were able to label hyoglossus motoneurons. However, in our pilot data, we were able to isolate styloglossus motoneurons effectively by reducing the volume of CTHRP injected, particularly in the older-age groups.
In the developing mammalian nervous system, most cells migrate from their location of origin to their final destination, where they reside permanently (Rakic, 1990). We thought this might be the case for the hyoglossus and styloglossus motoneurons during postnatal development. If neuronal migration occurred, it appears to be complete by the first week of life in the hyoglossus and styloglossus motoneurons. The lack of migration of postnatal motoneuron development found in this study is also consistent with postnatal development of genioglossus motoneurons within the ventral (protrusor) subcompartment of the hypoglossal nucleus of rat (Mazza et al., 1992; Sokoloff, 1993; Núñez-Abades et al., 1994) and kitten (Brozanski et al., 1989). Therefore, it appears that the motoneuron cell body translocation as a result of this neuronal migration is complete by the first week of life in the rodent hypoglossal nucleus.
The most rapid changes in cell size for hyoglossus and styloglossus motoneurons as measured by cross-sectional area appear to occur by the end of the second week of life. However, we noted a differential growth rate between hyoglossus and styloglossus motoneurons. Brozanski et al. (1989) found significant differences in cell body dimensions as measured by average diameter between 3–4 days, 2 weeks, and 4 weeks in genioglossus motoneurons of the cat. The greatest change occurred during the first 2 weeks of postnatal life, which is consistent with our data. Conversely, Núñez-Abades and Cameron (1995) did not find any changes in cell body diameter during the first 30 days of postnatal development in rat. However, due to the complexity of their study, the sample size was restricted to 10 genioglossus motoneurons per age group with four age groups being evaluated.
Cell roundness, as measured by form factor, progressively decreased with advancing postnatal age for both hyoglossus and styloglossus motoneurons. This finding is consistent with the trend of developing postnatal genioglossus motoneuron roundness as measured by form factor (Brozanski et al., 1989), which suggests that the shape of hyoglossus and styloglossus motoneurons is changing during postnatal development and into adulthood. However, there were differences found between hyoglossus and styloglossus motoneurons. Because form factor was measured by tracing the base of the primary dendrites and soma, differences between hyoglossus and styloglossus motoneurons in form factor may indicate differences in the spatial distribution of their primary dendrites.
Differences in the direction of hyoglossus and styloglossus motoneuron dendritic projections suggest possible differences among their premotor inputs. Altschuler et al. (1994) reported the parvocellular and intermediate reticular nuclei and the ventral medullary reticular nucleus to be similar premotor structures with direct inputs to the hyoglossus and styloglossus motoneurons in the rat. They also reported that the medullary reticular field had direct inputs to hyoglossus motoneurons while styloglossus motoneurons had direct inputs coming from the nucleus tractus solitarii. However, they found an intermingling of hyoglossus and styloglossus motoneurons within the hypoglossal nucleus, which is inconsistent with our data. In the present study, the direction of dendritic projection of these motoneurons was based on our observations. Future studies should utilize a quantitative measurement to elucidate differences in dendritic projection between hyoglossus and styloglossus motoneuron pools.
In summary, hyoglossus and styloglossus motoneurons are different in their hypoglossal nucleus location, cell sizes, cell roundness, and dendritic projections. Furthermore, the development of genioglossus (protrusor) motoneurons appears to be different from the postnatal development of retrusor motoneurons in the rate of cell growth (cross-sectional area) and rate of change in cell roundness when comparing to the same species. Interestingly, the postnatal development of retrusor motoneurons in the rat appears to be similar to the protrusor postnatal development in the kitten. This suggests that using genioglossus motoneuron development as a marker of hypoglossal development may not be all-inclusive. Future studies in our laboratory will evaluate morphological and physiological changes that occur during postnatal development of artificially reared rats, which are temporarily deprived of suckling activity.
Human infants born prematurely often have problems with suckling. As a result, they do not get the adequate nutrition they require to grow and develop normally. When this happens, the infants are given nutritional support (fed by alternate means such as nasogastric cannula and percutaneous endoscopic gastrostomy tube). Depending on the severity of the problem, the infant can go extended periods of time without nutritive suckling. Studies of other motor systems using an immobilization model have noted some neuromuscular changes (Booth, 1982; Gardiner et al., 1992; Jaweed et al., 1995; Picquet et al., 1998; Hoyer et al., 2000). While nutritional support may not restrict tongue movement, it does drastically reduce tongue use as it pertains to nutritive suckling (Bayer et al., 1983). One study using this artificially reared model has shown differences in contractile characteristics of the styloglossus when compared to normal dam-reared rats (Kinirons et al., 2003). This coexistence of low birth weight and nonoral feeding has been associated with motor speech deficits (Jennische and Sedin, 1998, 1999) and longer hospital stays (Schanler et al., 1999). It seems that disruption of the normal development of the motor control system for suckling and tongue movement may play a vital role in these speech deficits. Since we have now established the normal postnatal development of tongue retrusor motoneurons in our rat model, we can begin to evaluate postnatal development of hyoglossus and styloglossus motoneurons of artificially reared rats.