Food consistency has been shown to exert an influence on the bony masticatory apparatus and associated musculature, including bone strain levels (Hylander,1979), cortical bone and cartilage proportions and properties (Beecher and Corruccini,1981; Bouvier and Hylander,1981,1982,1984,1996a,1996b; Beecher et al.,1983; Ravosa et al.,2006a, in press), and activity levels of the jaw-closing muscles (Weijs and Dantuma,1981; Langenbach and van Eijden,2001). Some studies have further shown that dietary-induced alterations in muscle activity are associated with changes in fiber size and composition, particularly during ontogeny (Kiliaridis et al.,1988; Chang et al.,1995; Saito et al.,2004). Others have demonstrated alterations in fiber type cross-sectional area, but with no attendant changes in fiber composition (e.g., Langenbach et al.,2003). Although dietary consistency has been shown to influence fiber type composition of the masticatory muscles (e.g., Kiliaridis et al.,1988) and the cross-sectional area of slow-contracting fibers in the posterior deep masseter (Langenbach et al.,2003), little is known about the effects of dietary consistency on jaw muscle fiber architecture.
Muscle fiber architecture is an important determinant of the contractile properties of whole muscle (Gans and Bock,1965; Muhl,1982; Sacks and Roy,1982; Lieber et al.,1992; van Eijden et al.,1997; Taylor and Vinyard,2004,in press). Furthermore, it is well appreciated that the jaw-closing muscles in humans and other mammals exhibit inter- and intramuscular variation in mass, fiber orientation, and internal architecture (Gaspard et al.,1973; Herring et al.,1979; Weijs and Dantuma,1981; Bredman et al.,1991; van Eijden et al.,1997; Antón,1999). In rabbit masseter, the physiological cross-sectional area (PCSA) of the lateral portion of the superficial masseter is three and a half times greater than the PCSA of the medial portion, and more than six times greater than the anterior deep masseter muscle (Weijs and Dantuma,1981). Likewise, activity patterns of the rabbit chewing muscles show inter- and intramuscular heterogeneity. For example, during the power stroke of mastication, the two superficial portions of the balancing-side superficial masseter begin firing, and peak, prior to the deep portion of the superficial masseter (Weijs and Dantuma,1981). In light of this structural and functional heterogeneity, an appreciation for how muscle compartments respond to long-term alterations in dietary consistency is important, as such alterations ultimately impact on how the mandible is loaded during mastication and incision.
In this study, we explore the effects of variation in dietary consistency on the internal architecture of rabbit masseter. In addition, we assess whether alterations in fiber architecture occur homogeneously across various portions of the superficial masseter muscle. In rats (Hurov et al.,1986) and rabbits (Weijs et al.,1987), the jaw-closing musculature increases in size during growth relative to the jaw-opening muscles, and the superficial masseter comprises a substantially larger percentage of the total jaw-closing muscle weight in adults (Schumacher and Rehmer,1960; Turnbull,1970; Weijs and Dantuma,1975,1981; Weijs et al.,1987), affirming the importance of the superficial masseter for generating bite force [as well as facilitating jaw opening, e.g., Taylor and Vinyard (2004)]. Therefore, the null hypothesis is that lateral and medial portions of the superficial masseter do not differ in their response to dietary consistency. The alternative hypothesis is that the masseter shows regional (intramuscular) variation in response to differences in dietary consistency.
MATERIALS AND METHODS
Twenty genetically similar male and female New Zealand domestic white rabbits (Oryctolagus cuniculus) were randomly separated into two equal dietary groups of 10 each. To ensure similarity in postnatal response to dietary modification, and thus control for variation in genetics and phylogeny, only siblings were selected. All rabbits were obtained as 4-week-old healthy weanlings and kept in the AALAC-accredited Center for Comparative Medicine (CCM; Northwestern University Feinberg School of Medicine) under an IACUC-approved protocol.
Two dietary cohorts of 10 rabbits each were entered into the experimental protocol at weaning (4 weeks old) (Sorensen et al.,1968; Yardin,1974). The twofold advantage of this early timing is that it limits the postnatal effects of other dietary inputs into the jaw adductor muscle system, thereby improving the likelihood that any morphological differences can be reliably ascribed to variation in food material properties; and that it optimizes the timing of the experimental protocol to coincide with the postweaning period of marked ontogenetic plasticity (Bouvier,1988; Bouvier and Hylander,1996a,1996b; Ravosa et al.,2006a, in press). One group of weanlings was raised on a diet of ground rabbit pellets previously soaked in water (Harlan TekLad Rabbit Pellets) to model underuse (UU) of the masticatory complex, while the second group was fed a tough diet of intact pellets supplemented daily with two 2 cm hay blocks to model overuse (OU) (Ravosa et al.,2006a, in press). The inclusion of pellets in the diet of all postweaning rabbits ensures adequate nutrition for normal growth (cf. Chang et al.,1995). Rabbits were housed in plastic cages to reduce paramasticatory behaviors such as incisor wire gnawing.
Weanlings were raised on the UU or OU diets for a period of 105 days and sacrificed as subadults (19 weeks old). The skull was detached en masse at the vertebral column, which facilitated fixation of the jaw muscles with the mandible fully adducted. In this single-blind analysis, muscles were coded such that the investigator conducting the muscle fiber architecture analysis (A.B.T.) was blind to the dietary status of the rabbits (M.J.R.).
Fiber Architecture Data Collection
Following fixation (10% buffered formalin), the left masseter muscles were carefully dissected free of their bony attachments, trimmed of excess tendon and fascia, blotted dry, and weighed to the nearest 0.0001 g (Mettler-Toledo AB204-S). The masseter muscle was sectioned along the length of the muscle belly into superior, middle, and inferior segments (Fig. 1a). From these segments, three portions of the superficial masseter muscle were identified (Fig. 1b) (Weijs and Dantuma,1981).
Fiber length and angle of pinnation for the masseter were measured in each portion following Taylor and Vinyard (2004). Specifically, a muscle segment was oriented on its perpendicular to expose the individual fasciculi and their proximal and distal attachments to tendon. Each segment was pinned to a styrofoam block and placed under a five-diopter magnifier lamp. Anterior and posterior sampling sites were chosen along the length of each muscle segment corresponding to the three muscle portions. Anterior fibers were sampled consecutively by choosing a fiber from the anteriormost portion of the myotendinous junction (MTJ) and proceeding posteriorly, and posterior fibers were sampled in reverse fashion beginning at the posterior portion of the MTJ and proceeding anteriorly (Fig. 1b). At each sampling site, a maximum of 12 adjacent fasciculi was measured.
Measurements taken for each fasciculus include fasciculus length, between the proximal and the distal myotendinous junctions (lf); and the perpendicular distance from the tendon of insertion to the proximal attachment of the fasciculus (a). The angle of pinnation (θ) was calculated as the arcsin of a/lf (Fig. 2). Using the aforementioned measurements, the following architectural variables were computed for each muscle compartment, and for the whole masseter muscle, separately by diet group.
Mean fiber length for each muscle compartment was calculated as the average fiber length of the compartmental anterior and posterior fasciculi, respectively. Mean fiber length for whole masseter was calculated as the average of all fiber lengths from the superficial compartments along with fibers sampled from the deep masseter.
Physiological cross-sectional area by muscle compartment and for whole masseter was calculated as follows: PCSA (cm2) = [muscle mass (g) × cosθ]/[lf (cm) × 1.0564 g/cm3], where 1.0564 g/cm3 is the specific density of muscle (Murphy and Beardsley,1974).
Assumptions of normality were met based on the Shapiro-Wilk test, justifying the use of parametric statistical analysis. Because of possible sex differences in muscle mass (English et al.,1999), a one-way factorial analysis of variance was performed to test for significant group, sex, and interaction effects. Based on an a priori significance level of P < 0.05, results revealed significant group effects (P = 0.002), but no significant sex or interaction effects (P > 0.05). Therefore, the sexes were combined and Student's t-tests were used to assess whether the OU diet and UU diet groups differ significantly in masseter fiber architectural variables. Prior research (e.g., Langenbach et al.,2003) leads to the expectation that rabbits fed harder foods should demonstrate an increase in whole masseter PCSA. Therefore, a one-tailed test was performed for this analysis. However, because there is little evidence for or against the hypothesis that fiber architecture differences should be exp ressed uniformly across all portions of the superficial masseter muscle, two-tailed tests were determined to provide the most suitable approach to assess regional differences between groups.
Behavioral analysis and observation of a pilot sample fed ad lib indicate that OU diet rabbits regularly ingested a daily supply of two hay blocks; UU diet rabbits did not exhibit failure to thrive, nor did they develop incisor malocclusions; and 90% of the UU diet sample falls within the skull length range for 10 similar-aged OU diet rabbits. Average muscle mass is 6.75 g ± 0.79 and 5.35 g ± 0.89 for the OU diet and UU diet rabbits, respectively. Whole muscle mass is significantly greater in the OU diet rabbits (df = 1,17; F = 12.90; P = 0.002). On the other hand, there are no group differences in fiber length or pinnation angle (Table 1). The OU diet rabbits demonstrate a significant difference in whole muscle PCSA and in the PCSAs of the lateral and medial portions of the superficial masseter (Fig. 3, Table 1). The percentage difference in PCSA ranges between 16.0% and 20.0%, with the greatest percentage differences accruing to the deeper portions of the superficial masseter (superficial masseter portions 2 and 3).
Table 1. Means, standard deviations (SD), and results of statistical tests for differences in fiber architectural variables between the overuse diet (OU) and underuse diet (UU) rabbits.a,b,c,d
One muscle in the OU-diet group was damaged and was therefore excluded from data analysis.
Lateral masseter 1 represents the superficial masseter fibers running from the myotendinous junction to the zygomatic arch; Lateral masseter 2 represents the superficial masseter fibers running between the superficial and deep myotendinous junctions; Medial masseter 3 represents the fibers running from the deep myotendinous junction to the mandible. Symbols designating each portion of the superficial masseter are depicted in Figure 1b, following Weijs and Dantuma (1981).
Boldfaced values indicate significant differences (P ≤ 0.01) in PCSA between groups. No other differences were significant.
Lateral masseter 1 (○)
Lateral masseter 2 (▵)
Medial masseter 3 (•)
Results of this study show that variation in food consistency produces alterations in the masseter muscle. Postweaning rabbits fed whole pellets and hay blocks showed an increase in PCSA compared to rabbits raised on a diet of crushed pellets alone. Pellets are hard and brittle, whereas hay is tough (Ravosa et al.,in press). Both foods require the generation of large muscle and bite forces, but whole pellets and hay entail more repetitive loading of the jaws to process than pellets soaked in water and pulverized. Thus, the increase in PCSA observed in the OU diet rabbits likely derives from the combined effect of both large and repetitive jaw loads.
The differences in PCSA are largely a function of increase in muscle mass, as average masseter muscle weights differ significantly, but fiber lengths and pinnation angles do not. Findings from this study are consistent with previous work demonstrating that variation in dietary consistency induces significant changes in relative masseter muscle mass (Kiliaridis et al.,1988; Chang et al.,1995) and fiber cross-sectional area (Langenbach et al.,2003). Average masseter fiber lengths for the subadult OU diet rabbits, though slightly smaller, are close to values reported by previous investigators for normal-fed adult rabbits (e.g., Schumacher and Rehmer,1960; Weijs and Dantuma,1981). Any inconsistencies in absolute values among these studies are likely accounted for by differences in developmental age at the time of measurement.
The OU-fed rabbits exhibit an increase in PCSA of the superficial masseter compared to the UU-fed rabbits (Table 1). This increase is observed in all portions of the superficial masseter muscle. Thus, despite functional heterogeneity in the masseter muscle (e.g., the balancing-side superficial portions fire first, followed by the deeper portions) (Weijs and Dantuma,1981), larger muscle and bite forces are required throughout both the superficial (Fig. 3, Table 1) and deep (e.g., Langenbach et al.,2003) masseter in order to generate adequate bite forces for mastication of a tougher diet. These findings are also empirically congruent with previous experimental study (Weijs and Dantuma,1981) demonstrating that in rabbit masseter, all muscle portions (both superficial and deep) reach maximum activity levels during ipsilateral chewing of hay compared to mastication of pellets or carrots.
The differential increase in muscle weight and PCSA over fiber length corresponds with previous studies demonstrating that in mammalian pinnate-fibered muscles, postnatal growth is characterized by increases in muscle width, either by increasing myofibril number, diameter, or both, while longitudinal growth is completed early in ontogeny (Carlson,1983; Stickland,1983; Herring and Wineski,1986; Woittiez et al.,1986; Langenbach and Weijs,1990). Weijs and Dantuma (1987) observed a 15% increase in fiber length from weanling to adult rabbits, compared to a fivefold increase in muscle weight during the same period of growth. Although increase in pinnation angle can improve muscle (and hence bite) force (Herring et al.,1979; Gans,1982), results of this study suggest that augmenting pinnation angle is not a biomechanical solution to improving muscle force during growth, as the OU-fed rabbits demonstrated an increase in PCSA solely as a function of increase in masseter muscle mass. The finding that OU and UU of the masticatory complex during growth largely influences muscle mass suggests that alterations in dietary consistency have an impact on masticatory force production, but not necessarily on jaw movement or excursion. By contrast, increase in masseter and temporalis muscle fiber length, accompanied by a reduction in pinnation angle, has been associated with requirements for increased muscle stretch in animals generating wide jaw gapes during feeding (Taylor and Vinyard,2004,in press; Eng et al.,2005). However, the extent to which these observed differences in fiber length occur during growth remains unknown.
It has been previously demonstrated (Langenbach et al.,2003) that dietary consistency influences the cross-sectional area of posterior deep masseter fibers, but not of superficial masseter fibers. However, in the Langenbach et al. (2003) study, animals were entered into the experimental dietary protocol as subadults (∼16 weeks of age). Therefore, the preexperimental diet may have influenced masseter muscle growth (e.g., Weijs et al.,1987), so that the extent to which differences between the two rabbit groups can be attributed to variation in dietary consistency remains somewhat ambiguous. Moreover, age-related decreases in plasticity have been shown to characterize bony, cartilaginous, and muscular tissues of the masticatory and other systems (Bouvier,1988; Chang et al.,1995; Bouvier and Hylander,1996a,1996b; Welle et al.,2000; Narici et al.,2004; Kim et al.,2005; Ravosa et al.,2006a, in press). Indeed, differences in the timing of implementation of experimental protocols may account in part for the disparity in findings between the Langenbach et al. (2003) study, which showed no changes in fiber composition, and those of other investigators who initiated specialized diets soon after weaning (e.g., Kiliaridis et al.,1988). Although beyond the scope of this study, the observed increase in PCSA of the superficial masseter suggests it would be worthwhile to reevaluate the influence of dietary consistency on masseter fiber composition early in ontogeny (i.e., at or near weaning).
In summary, it can be concluded that dietary consistency influences rabbit masseter PCSA, but not fiber length or pinnation angle. Specifically, rabbits masticating tougher foods are characterized by an increase in masseter PCSA, largely accounted for by increase in muscle mass, compared to genetically similar rabbits masticating a less tough diet. Moreover, all portions of the superficial masseter muscle exhibit increases in PCSA, indicating that improved muscle and bite force occurs homogeneously throughout the superficial masseter. Results presented here suggest that alterations in muscle fiber architecture are facilitated during early growth due to variation in dietary consistency.
Supported by the Department of Cell and Molecular Biology, Northwestern University (to M.J.R.) and the National Science Foundation (BCS-0412153; to A.B.T.). Barth Wright kindly performed the analyses of rabbit food material properties. Two anonymous reviewers provided useful comments that improved the quality of this manuscript.