Role of the osteoclast in cranial suture ... - Wiley Online Library

0 downloads 0 Views 711KB Size Report
This study investigates the role of bone resorption in defining inter- digitations characteristic of ... 2006 Wiley-Liss, Inc. Key words: osteoclasts; cranial sutures; fractal geometry; alen- .... search identifies cyclical (masticatory), rather than static. (brain growth) ..... element analysis (FEA) in order to model a suture wave-. Fig. 5.
THE ANATOMICAL RECORD PART A 288A:552–563 (2006)

Role of the Osteoclast in Cranial Suture Waveform Patterning CRAIG DANIEL BYRON* Department of Surgery, Medical College of Georgia, Augusta, Georgia

ABSTRACT This study investigates the role of bone resorption in defining interdigitations characteristic of cranial suture waveform. Male mice from the CD-1 (ICR) background were analyzed at six age groups (n ⫽ 5 mice per group) in order to study the ontogenetic changes of osteoclast counts using tartrate-resistant acid phosphatase-stained histological sections of sagittal sutures. Additionally, the complexity of suture lines were measured ectocranially from the same age groups (n ⫽ 5 per group) using image capture and fractal geometry (ruler dimension method). The results suggest that osteoclast resorption is a contributor to suture patterning. Specifically, osteoclasts show the greatest activity along concave suture regions at 42 and 84 days (Kruskal-Wallis test statistic ⫽ 14.9; P ⱕ 0.01). This coincides with significant increases incrementally in suture complexity as measured with fractal dimension at 42 and 84 days of age (ANOVA F-statistic ⫽ 19.84; P ⱕ 0.001). In congruence with these data, mice given osteoclast-depleting injections of alendronate show a decrease in sagittal suture complexity. Data from this experiment indicate a positive relationship between suture complexity and osteoclast count (P ⬍ 0.01; r ⫽ 76%). Increases in suture complexity and osteoclast activity occur after peak rates of cranial width growth and coincide with weaning and the transition to a hard chow diet. These data demonstrate osteoclasts along the bone margin of the cranial suture and also indicate that sutures attain their complex shape at the same age when osteoclast number is highest along concave suture margins, underscoring the role of osteoclasts in generating the suture waveform pattern. Anat Rec Part A 288A:552–563, 2006. © 2006 Wiley-Liss, Inc.

Key words: osteoclasts; cranial sutures; fractal geometry; alendronate; pattern formation

Cranial sutures connect interfacing flat bones of the skull vault and face and arise developmentally as amorphous cellular/fibrous connective tissue ligaments. Characteristically, these sutures become patterned as a waveform demonstrating interdigitating or interlocking bony projections when viewed from an ectocranial perspective. Appositional bone growth along these suture margins is primarily responsible for the lengthening of calvarial bones rather than their thickening, i.e., the cranial suture is to calvarial growth as the epiphyseal growth plate is to long bone growth (Wolf et al., 1985a; Mao, 2002, 2005). The coordinated longitudinal growth of cranial bone is important for housing developing sensory structures involved in vision, olfaction, and audition. Most importantly, proper bone growth in the calvarium is required for normal growth of the brain. Improper bone formation along cranial sutures commonly manifests as sutural bony fusion, i.e., craniosynostosis. This pathology restricts cranial ©

2006 WILEY-LISS, INC.

vault growth along a single axis or along multiple axes depending on how many calvarial sutures are affected. This condition that presents in roughly 1 in 2,500 human births can be responsible for innumerable craniofacial deformities (Cohen, 1997; Greenwald et al., 2000; Opperman, 2000).

Grant sponsor: U.S. National Science Foundation; Grant number: BCS-0352512. *Correspondence to: Craig D. Byron, Department of Surgery, CJ-1117, Medical College of Georgia, 1120 15th Street, Augusta, GA 30912. Fax: 706-721-6931. E-mail: [email protected] Received 6 October 2005; Accepted 16 December 2005 DOI 10.1002/ar.a.20322 Published online 7 April 2006 in Wiley InterScience (www.interscience.wiley.com).

ONTOGENETIC CHANGES IN SUTURE OSTEOCLASTS

The vertebrate cranium exists as a highly complex and integrated structure. Moss’s functional matrix hypothesis predicts that cranial skeletal tissues are modeled by the additive functional influences of all cranial soft tissues on the osteogenic environment (Moss, 1997a, 1997b, 1997c, 1997d). In the case of the skull, an important osteogenic environment to be considered is the cranial suture. The regulation of cranial suture osteogenesis includes both autonomous molecular, as well as functional physical controls (Enlow, 1990; Mao, 2005). The biomechanical events that influence suture bone growth can be considered here twofold. First, the need to expand the calvarium to keep pace with encephalization inarguably plays an important role via molecular signaling from the cranial meninges. Suture cells exist in a tightly controlled cycle of proliferation, transdifferentiation/migration, and cell death. Orchestration of these events is in large part by the dura mater, as revealed by Opperman et al. (1993, 1995) and Winograd et al. (1997) in key experiments demonstrating fusion in fully formed sutures with the absence of dura. Additional experiments reviewed by Opperman (2000) revealed that the dura secretes a variety of growth and transcription factors that regulate cell activity within the overlying suture. These include TGF-␤1, TGF-␤2, TGF␤3, FGF-2, BMP-4, BMP-7, FGF-9, MSX1, and MSX2. Because suture fusion occurs within certain sutures as a normal sequence in cranial vault growth, the ability of the dura to secrete these factors is hypothesized to differ depending on vault location and ontogenetic timing (Greenwald et al., 2000). A second suite of functional events influencing suture growth patterning arise from external biomechanical loads that are dissipated throughout the craniofacial architecture. These are mainly considered here as masticatory loads due to the contraction of paired primary chewing muscles such as masseter, temporalis, and pterygoid that attach to cranial elements. The resection of these muscles has been shown to diminish bone formation, thereby distorting craniofacial growth (Moss, 1957, 1961; Wolf et al., 1985b). It is recognized that significant loads may arise from the behavioral use of the skull. This is especially pertinent to consider when the skull has bony appendages that are used to grapple and duel with competitors such as is broadly observed within the mammalian order Artiodactyla. Cranial suture waveform pattern, or complexity, of vault sutures has frequently been related to mechanical loads resulting from mastication (Behrents et al., 1978; Steenvoorden et al., 1990; Katsaros et al., 1994; Moss, 1997a; Kobayashi et al., 1999; Kopher and Mao, 2003; Byron et al., 2004). If true, then one would expect agerelated changes in suture complexity to occur as an individual grows in size and places greater loads on its masticatory apparatus. In this regard, the quantification of ontogenetic suture complexity is incomplete. Previously, we have suggested a model for sagittal suture growth whereby bone formation occurs from tensional stress along convex suture lingulae and is enhanced in mice with significantly increased temporalis muscle mass (Byron et al., 2004). The iteration of this pattern through time resulted in a more complex and interdigitating suture line in hypermuscular individuals as measured with fractal geometry. The observation that bone formation occurs on convex surfaces and contributes to suture complexity begs the following question. What role does bone resorption

553

play along concave vs. convex margins and how might this contribute to suture waveform patterning? Osteoclasts have been implicated in modeling cranial bone thickening (along the endoectocranial axis) as well as lengthening (Rice et al., 1997, 1999; Winograd et al., 1997, 2001). Controlling the osteoclasts along the lengthening calvarial bone margins is likely involved in premature suture fusion. Despite the claims by Burke et al. (1995) that osteoclasts are only found in the diploic spaces of flat bones and Vastardis et al. (2004) that osteoclasts are not involved in unilateral coronal synostosis, other evidence (including research presented here) disagrees. The osteopetrotic mouse (op/op) lacks osteoclast cells due to a macrophage colony-stimulating factor deficiency. These mice demonstrate immature bony fusion of cranial sutures, suggesting the important role osteoclasts play in maintaining suture patency (Kawata et al., 1998; Kaku et al., 1999). Several pharmaceutical agents are commercially available for negatively regulating mature osteoclast populations. Bisphosphonates are one of these classes of drugs commonly used to treat osteoporosis (excessive bone resorption). These agents adsorb to mineralized surfaces within bone tissue. Osteoclasts, acting as bone tissue macrophages, resorb these surfaces through routine bone turnover. In this way, bisphosphonates access a macrophage’s “interior” and disrupt vital metabolic processes (Carano et al., 1990; Baron, 1999; Green and Clezardin, 2002). Surprisingly few studies exist that utilize these agents to manipulate bone apposition in the calvarium. Interestingly, an interpretation was made by Kawata et al. (1998) that osteoclast deficiency accompanied by decreased masticatory force associated in the same animals may lead to craniosynostosis. Not only does this implicate masticatory loads in cranial suture biology, but it may in part validate the microspicule hypothesis of Burke et al. (1995) that mechanical loads are necessary to fracture newly formed microspicules across the suture. Other research identifies cyclical (masticatory), rather than static (brain growth), mechanical strain to be a factor modulating suture growth (Rafferty and Herring, 1999; Herring and Teng, 2000; Kopher and Mao, 2003; Sun et al., 2004). These data emphasize the importance of masticatory function rather than expansional brain growth in the mechanics of sutural growth. The role brain growth plays in suture development is further diminished by observations about rapid rates of bone apposition in the calvarium in early postnatal life. Specifically, bone apposition in the rat calvarium observed by Massler and Schour (1951) at 100 ␮m/day cannot be explained by rapidly diminishing static strain levels emanating from an expanding brain calculated using suture material properties (Henderson et al., 2004). Fong et al. (2003b) provide intracranial volume data (ICV) for rats suggesting that brain growth is roughly complete by 60 days postnatal. Daily strain levels from the growing brain were calculated to diminish rapidly after birth such that by 10 days strain was only around 2%. In a separate in vitro study of calvarial osteoblasts, Fong et al. (2003a) showed that 10% equibiaxial strains were sufficient to illicit an appropriate mechanobiological response in calvarial osteoblasts. They also showed that 2.5% equibiaxial tensional strain does not elicit a morphogenetic response. Despite the rapid early growth of the brain in the rat model, tensile strains simply diminish far earlier than calvarial lengthening. It seems more likely that mechan-

554

BYRON

ical strains sufficient to drive sutural morphologic patterning emanate from mechanical loads delivered via the masticatory muscles. The purpose of this study was to test the hypothesis that suture complexity increases with age in a mouse model and that bone resorption along concave suture margins by osteoclasts participates in this patterning. These hypotheses will be addressed using a combined morphometric and histological approach in an ontogenetic series of mice as well as in bisphosphonate-treated vs. control age groups. Results from this analysis will be presented in a manner to address the participation of brain growth and feeding mechanics in this ontogenetic suture patterning.

MATERIALS AND METHODS Mice In the ontogenetic series, six groups of mice (n ⫽ 10 per group) were examined at ages representing birth to full maturity. Postnatal CD-1 (ICR) mice at 1, 10, 21, 42, 84, and 168 days were obtained from Charles River Laboratories (Wilmington, MA). This mouse strain is representative of a standard outbred albino mouse and was selected for no specific reason other than its use in previous studies by the author. The six age groups between 1 day and 6 months were chosen since this time period broadly captures postnatal development from infancy to adulthood. The age groups were spaced such that each group is approximately twice as old as the preceding age group. This pattern allows more ages to be evenly sampled during early postnatal development when skeletal growth is most rapid. Each age group was divided in half such that five animals were used for histochemistry and the other five were used for fractal analysis. Additionally, two groups of mice (n ⫽ 10 per group) were given alendronate injections or control (PBS) injections weekly for 3 weeks following weaning at 21 days. Postnatal 21 day CD-1 (ICR) mice for this osteoclast knockdown experiment were purchased from the transgenic mouse/stem cell core facility at the Medical College of Georgia. These mice are from the same strain as those used in the ontogenetic series and were purchased from the author’s institutional core facility rather than Charles River due to the immediacy of transfer and the desire to begin injections at the same time when animals were weaned. All mice were euthanized with a CO2 overdose followed by a thoracotomy according to the guidelines of the sponsoring institution (AUP 03-08-315).

Histochemistry Sagittal sutures from 5 mice within each age group in the ontogenetic series and 10 mice from each group (experimental vs. controls) in the osteoclast knockdown experiment were excised with attached parietal bone bordering either side. These specimens were cleaned of dura mater and periosteum, fixed in 4% paraformaldehyde, and decalcified in 4% EDTA. The sutures were then embedded in paraffin, cut in a transverse plane 4 ␮m thick, and mounted on glass slides. Slides were then stained for tartrate-resistant acid phosphatase (TRAP) with a kit purchased from Sigma-Aldrich (386A-1KT). This kit is designed to demonstrate acid phosphatase activity in leucocytes from blood and bone marrow. It is a good assay for osteoclasts since they are bone marrow macrophages that utilize acid phosphatase to degrade bone extracellular ma-

trix. TRAP-stained suture sections were then digitally imaged with a Leica compound light microscope equipped with a camera and computer interface. The sagittal suture was photographed at 20⫻ in its entire x-y plane of view. These individual frames were fit together in Adobe Photoshop 6.0 Professional so that each individual suture could be viewed in one image with extremely high resolution.

Complexity Analysis The entire cranium from five mice within each age group were removed. These specimens were cleared and stained following Wassersug (1976). This technique uses KOH to macerate soft tissue, alcian blue to stain cartilage, and alazarin red to stain bone. Completed specimens are stored in glycine and reliably demonstrate skeletal tissue morphology. Cleared and stained heads were digitally imaged at 1.25⫻ using a Leica S6D stereoscopic dissecting microscope with digital camera and computer interface. Specimens used in the osteoclast knockdown experiment were photographed using contrasting light with the same microscope setup prior to these specimens’ processing for histological analysis. Digital traces of sagittal sutures were created using Adobe Photoshop 6.0 Professional and saved as BMP files. These files were then opened with Benoit 1.3 fractal analysis software (Trusoft International) and measured using the ruler dimension method. These same digital images were also measured using a relative length ratio (Jaslow, 1990; Jaslow and Biewener, 1995) in order to provide additional support for the quantification of suture morphology. This value measures the chord length between landmarks at the beginning and end of the suture. This line is then divided by the linear path length of the suture waveform. The result is a measure of suture complexity assessing path length as a percent of the chord length. Sutures with greater complexity will have a higher value. Sagittal sutures were chosen for analysis due to prior data collected on this calvarial suture (Byron et al., 2004) in addition to the demonstrated biomechanical milieu this region shows during temporalis contraction and mastication.

Alendronate Injections In the osteoclast knockdown experiment, 20 mice were randomly placed into one of two groups upon weaning at 21 days (3 weeks). An experimental group received weekly injections of alendronate (2.5 mg/kg/wk; Sigma Aldrich A4978) following Evans et al. (2003), while a control group received PBS. Alendronate was used due to its ability to adhere to bone mineral and kill mature osteoclasts upon resorption of that mineral (van Beek et al., 1997; Ma et al., 2003). Due to its properties as a bisphosphonate drug, alendronate effects are limited to the mineralized skeleton. No known adverse effects occur in soft tissue growth and development. Newly weaned mice were given injections over 3 weeks until they were aged to 6 weeks. At this time, the experiment was ended and sagittal sutures were harvested for morphometric and histologic analysis. The start and endpoints for administering alendronate injections were significant of developmental events. At 3 weeks, these mice were weaned from their mothers and given a standard hard chow that requires mastication. This starting point represents a dietary transition from low masticatory to higher masticatory mechanical load-

ONTOGENETIC CHANGES IN SUTURE OSTEOCLASTS

555

Fig. 1. A: Representative sagittal suture histosections (20⫻) stained for TRAP from each postnatal day age group. B: Outline of suture-parietal bone margins on P42 20⫻ sections that have been refit into one image. Labels correspond to suture margin regions (convex, concave, and straight) and serve as input for computing a concave difference value.

ing. By 6 weeks of age, sagittal sutures attain a majority of their complexity as well as demonstrate excessive osteoclast activity along concave suture margins.

Skull Width In order to address the issue of brain enlargement and its role in stimulating suture growth, a proximate measure was collected using calvarial width. This measure was collected on the cleared and stained crania used for the ontogenetic complexity analysis. Mitutoyo hand calipers were used to measure calvarial vault width from the ectocranial perspective to the nearest hundredth of a mil-

limeter. These data were normally distributed within each age group and an analysis of variance (ANOVA) design was employed to compare width differences between ages. These data are used to approximate the rate of brain growth in this sample by assessing how the cranial vault grows transversely and how this relates to observed suture patterning.

Quantification and Statistical Analysis In the ontogenetic series, osteoclasts were counted from whole TRAP-stained suture images with reference to their location, i.e., along convex, straight, or concave bone mar-

556

BYRON

gins (Fig. 1). A fourth variable concave difference (CD) was calculated by subtracting the amount of convex osteoclasts within a given suture from the amount of concave osteoclasts. Osteoclast counts between each age group were compared for convex, straight, and concave lingulae as well as comparing CD values using Statgraphics Plus 5.0. These frequency data were compared using the Kruskal-Wallis test, which is a nonparametric equivalent to ANOVA. This nonparametric analysis was chosen because these frequency counts demonstrate nonnormal variation, kurtosis, and skewness, thus they violate the assumptions of the ANOVA design. Concave difference is computed by subtracting one count from another. This operation does not generate normally distributed data and thus CD is also compared using the Kruskal-Wallis statistic. Differences between age groups within each suture region were considered significant at the 95% confidence interval. Within the osteoclast knockdown experiments, total osteoclast counts were measured for each individual within a single sagittal suture section. These counts were compared between groups using the Mann-Whitney Ustatistic, the nonparametric equivalent to Student’s t. Additionally, a measure of suture complexity for each individual was correlated with osteoclast count in order to observe the relationship between osteoclast number and suture complexity in this experiment. It is recognized that a degree of subjectivity exists when attempting to identify mature osteoclasts and classify them according to convex, straight, and concave regions. Figure 1 gives an example of a suture tracing with representative regions classified. Convexity vs. concavity was decided in how the margin relates to the long axis of the entire suture. Unless the concavity or convexity was matched by its interlocking side from across the suture, it was classified as straight. Osteoclasts by nature occupy hollowed-out depressions called Howship’s lacunae. These concave pits exist at an order of magnitude below the scale used to identify suture region along the suture path and therefore did not influence the assignment of a suture region. Additionally, osteoclasts exist in several states of development within bone. Rice et al. (1997) demonstrate that TRAP-positive osteoclasts tend to be associated with more mature bone mineral while MMP-9-positive osteoclasts are often associated with new bone mineral resulting from the rapid osteogenesis characteristic of early skeletogenesis. In this regard, neither TRAP nor MMP-9 is the perfect assay for observing osteoclasts across several ages. However, TRAP positivity was used in this analysis since most of the ages analyzed here occur after the skeleton is formed, when mature bone mineral is being resorbed. Figure 2 demonstrates criterion used to identify TRAP-positive osteoclasts. Multinucleated TRAP-positive cells that were associated with the bone margin, especially when residing in a pit, were interpreted as osteoclasts actively modeling the cranial suture. Fractal dimension is a measure of a shape’s self-similarity upon iterative magnification and has proven to be a successful alternative to traditional Euclidian geometry in the analysis of form in nature (Mandlebrot, 1967, 1977; Hartwig, 1991; Long and Long, 1992; Cross, 1997; Monteiro and Lessa, 2000; Lynnerup and Jacobsen, 2003). Fractal analysis enables complex planar shapes to be quantified by assigning them a unit-free measure that lies between one and two dimensions (1.0 –2.0). The ruler dimension method, also known as structured walk, calcu-

lates fractal dimension by using a ruler with a length that varies according to a constant value (coefficient of decrease) to measure the length of a jagged line. For this study, the largest ruler used was 600 pixels (approximately 1,800 ␮m) and the smallest ruler was 5 pixels (approximately 15 ␮m). These ruler dimensions were used so that the fractal analysis would begin at a scale approximate to the entire length of the sagittal suture. The fractal analysis proceeds by measuring suture length with increasing resolution ending with a measurement that appreciates the linearity of the suture at the level of a single cell. Generally with complex objects, overall length becomes longer as the ruler used to measure length becomes shorter. Ruler dimension (i.e., Dr or FD) is taken as the slope of a line where log ruler length is on the x-axis and log N of rulers (overall suture length) is on the y-axis. Dr is a scaling value relating changes in ruler length to changes in overall suture length. The steeper this slope is, the more complexity the suture demonstrates. A reliability test using these parameters on a straight line yielded an FD measurement of 1. Mean fractal dimensions were distributed such that ANOVA could be used to test for differences among age groups and Fisher’s LSD (least significant difference) could be used as a posterior test for differences between ages. Differences were considered significant at the 95% confidence interval.

RESULTS Osteoclast Localization Staining for osteoclasts using the TRAP protocol reveals significant differences between age groups in the amount of osteoclasts present. The raw counts for convex, straight, and concave regions (Fig. 3) are converted to ranks and show significant Kruskal-Wallis test statistics and P values (Table 1). This demonstrates there are significantly different amounts of osteoclasts between ages when considering each suture region individually (i.e., convex, concave, and straight). A common trend emerges when comparing convex, straight, and concave regions. Osteoclast number increases with age until around 42 days postnatal (6 weeks), after which osteoclasts decrease substantially in convex and concave regions. Osteoclasts along straight margins are most numerous at 21 days and then decrease according to this same general pattern. With these data, a fourth variable was calculated called concave difference. This simply represents the number of concave osteoclasts minus the number of convex osteoclasts per individual. These regions are depicted in Figure 1B. The CD variable represents the degree to which osteoclasts are localized to the concave regions relative to the convex regions within each age group. As can be seen from Figure 3 (red line) and Table 1 (column d), there is a significant difference among age groups with regard to the localization of osteoclasts along concave suture lingulae. Analysis of the CD line reveals that postnatal day 42 (6 weeks) shows the highest amount of concave relative to convex osteoclast activity. This increase is approximately double the count observed at P21 and occurred over the course of the 3 weeks that followed weaning.

Suture Complexity and Skull Width Using fractal geometry to evaluate the complexity of the sagittal suture in combination with osteoclast localization data allows for ontogenetic changes in bone resorption

ONTOGENETIC CHANGES IN SUTURE OSTEOCLASTS

557

Fig. 2. Iteratively magnified histological image of a TRAP-stained specimen. This outlines criterion used to classify mature osteoclasts. Note the multinucleation highlighted with arrows as well as resorption pits (Howship’s lacunae).

within the suture to be related to increases in suture amplitude. Table 2 provides the summary and ANOVA statistics comparing fractal dimension at each age group. Figure 4 illustrates these data using line plots. Sagittal suture fractal dimension (Fig. 4A) increases as mice age. Posterior testing using Fisher’s LSD at the 95% confidence interval reveals successive increases in fractal dimension at 6 weeks (p42) and 3 months (p84). Relative suture length values (Fig. 4B) for each age mirror this complexification. Both measures show a significant increase in suture complexity occurring between ages 21 days and 42 days. Figure 5 illustrates this ontogenetic increase in suture complexity by showing representative micrographs of mouse calvariae from each age group. In order to observe how these increases in sagittal suture complexity relate to changes in brain growth, skull width was measured for each age group to the nearest 0.01 mm. ANOVA testing

revealed an obvious significant trend toward increasing skull width (Fig. 6, Table 2) that is especially pronounced between postnatal days 1 and 10. Posterior testing revealed significant differences between all age groups except those at 6 weeks and 3 months. Note that roughly 80% of transverse cranial growth occurred by 10 days of age. This rate of growth is not matched in rapidity by suture complexity growth. However, Figure 7 shows there is a strong correlation between skull width and suture complexity. Using skull width as an approximation for brain growth allows for the consideration of the expanding brain as a covariant to suture complexity. This will be explored below.

Osteoclast Knockdown and Suture Complexity Six-week mice given 3 weeks’ worth of alendronate injections showed a modest decrease in osteoclast count. This was

558

BYRON

Fig. 3. Osteoclast counts are presented here as growth rates. Lines pass through each groups’ mean count value and the error bars in the y-axis represent the standard error. Note the red line (CD) is computed as the difference between concave and convex lines. As such, it is located between these growth lines and communicates information about both concave and convex bone margins.

TABLE 1. Summary statistics: Ranked osteoclast counts Mean Rank of Osteoclast Count per Region Age (days)

N⫽30

a: Convex

b: Straight

c: Concave

d: Concave Difference

8 8 24 21 18.6 13.4

7.5 12.1 26.9 20.5 13.9 12.1

4 11.7 21.7 23.9 20.7 11

6 15.1 17.6 22.6 22.1 9.6

*16.9

*17.6

*19.9

*14.9

1 5 10 5 21 5 42 5 84 5 168 5 Kruskal-Wallis Test Statistic⫽ * ⫽ P⬍0.01

TABLE 2. Summary statistics and post-hoc tests: Fractal dimension and skull width Age (days)

N⫽30

Mean FD

1 10 21 42 84 168

5 5 5 5 5 5

1.004 ⫾ 0.028 1.035 ⫾ 0.022 1.039 ⫾ 0.025 1.107 ⫾ 0.026 1.167 ⫾ 0.059 1.133 ⫾ 0.017

Groups using Fisher’s LSD (Pⱕ0.05)

Mean Skull Width (mm)

X X X X X

accompanied by a modest decrease in sagittal suture fractal dimension. The Mann-Whitney U-statistic was used to test for differences between each treatment group. Unfortunately, no significant differences were detected between the FD and osteoclast count means of alendronate vs. control groups (data not included). However, by correlating a mea-

X X

6.89 ⫾ 0.17 10.25 ⫾ 0.11 10.6 ⫾ 0.22 11.51 ⫾ 0.32 11.52 ⫾ 0.15 12.31 ⫾ 0.22

Groups using Fisher’s LSD (Pⱕ0.05) X X X X X X

sure of suture complexity (relative suture length) onto osteoclast count for each individual, a significant positive relationship emerges between osteoclasts and suture complexity (Fig. 8; r ⫽ 76%; P ⬍ 0.01). Figure 9 compares typical sagittal suture sections from the alendronate (Fig. 9A) and control (Fig. 9B) groups.

ONTOGENETIC CHANGES IN SUTURE OSTEOCLASTS

Fig. 4. Line plots demonstrate increasing sagittal suture complexity with age in this mouse model. A: Fractal dimension as a measure of complexity. B: Relative length as a measure of complexity. The line passes through the mean for each group. Error bars in the y-axis show the standard error of the mean.

DISCUSSION Osteoclasts demonstrate ontogenetic changes in site specificity. Figure 3 and Table 1 clearly show that TRAPpositive osteoclasts are absent during initial postnatal development. This does not preclude the presence of osteoclasts at this time. In fact, Rice et al. (1997) have shown MMP-9-positive osteoclasts at these ages. The probable reason for this is because the calvarial bone being deposited at these newborn stages is low in mineral content. The matrix metalloproteinase, MMP-9, is thought to be sufficient for the early requirements of bone resorption. Later in postnatal development as bone becomes more densely mineralized, TRAP-positive osteoclasts would be required for resorption. Congruent with this explanation, osteoclasts are observed along concave and straight sagittal suture margins at 10 days postnatal. By 21 days postnatal, one can observe osteoclasts along convex margins as well. The occurrence of osteoclasts along convex and concave regions increases incrementally until 42 days postnatal and then incrementally decreases, creating a bell-shaped distribution ontogenetically. Osteoclasts along straight margins follow a very similar pattern except that the maximum number of straight osteoclasts is observed at 21 days postnatal. This is possibly related to the fact that there are more straight regions in younger, less complex sutures on which osteoclasts may develop. After 21 days, there is a significant increase in suture complexity (Fig. 4) and therefore fewer straight suture margins. These raw osteoclast counts reveal that bone resorption reaches its climax at 6 weeks of age. By 6 months, osteoclast activity is diminished to levels seen only very early in development around 10 days postnatal. Late in life after the mouse skeleton matures, there is likely to be less bone remodeling since the mechanical demands on these elements has plateaued.

559

By calculating concave difference, one is able to compare how concave osteoclasts relative to convex osteoclasts compare at different ages. Generally, one observes a very similar pattern as the nontransformed counts per region (Fig. 3, red line). In other words, there is a wax-and-wane shape reflecting an increase in concave osteoclasts relative to convex osteoclasts starting at 10 days postnatal and continuing until 42 days postnatal. After this age (6 weeks), there is a decline in osteoclasts along the concave relative to convex margins. When considering these results together with ontogenetic suture complexity data (Fig. 4) and osteoclast knockdown data (Fig. 8), it seems clear that osteoclast activity is involved in sagittal suture patterning. Evidence of this can be found by examining the marked increase in suture complexity between 21 and 42 days in both fractal dimension and relative length measures. This corresponds with concave osteoclasts most severely outpacing convex osteoclasts at 42 days. Alendronate-injected mice also showed decreased osteoclast counts and sagittal suture complexity, although these differences were not significant at the 95% confidence interval. There was a significant positive relationship observed between suture complexity and osteoclast count (Fig. 8) within alendronate and control mice that support the argument of a mechanistic role for osteoclasts in suture morphology. It is likely that alendronate injections carried out into adulthood (4 months) in these mice would generate further differentiation between treatment groups along the x- and y-axes. In this study, the experiment was ended at 6 weeks because data from Figures 3 and 4 suggested this 3-week time period (between 3 and 6 weeks) to be pivotal in cranial suture morphogenesis via osteoclasts. The ontogenetic pattern of osteoclast localization (Fig. 3) does not need to match the upward sloping pattern of fractal dimension (Fig. 4) as long as the rate of suture bone growth slows with age. If suture bone growth did not slow after 6 weeks, then one could expect bony suture fusion since bone resorption is relatively absent at 6 months. However, sagittal sutures in mice do not fuse with age. Moreover, Figure 4 demonstrates that complexity continues to increase with age. One potential explanation for the early activity of osteoclasts when suture complexity is low, is that suture bone growth is rapid below 6 weeks of age. To counteract suture obliteration and craniosynostosis, osteoclasts show more activity across all suture regions at these early ages. The stimulus for growth during these early stages may be mechanical. Figure 6 shows skull widths for mice at each age interval. The 80% of growth that occurred between ages 1 and 10 days is likely due to the rapidly growing brain. It is interesting to note that in rodents, brain growth is complete by 60 days postnatal (Fong et al., 2003b). The increases in suture complexity observed below 42 days could be precipitated by a mechanical signal such as tension due to brain growth. This could help explain why osteoclasts increase in activity along all regions at 21 days; they are responding to brain growth-related mechanical strain. However, it is the view of this author that after 3 weeks there is a discrete onset of mechanical forces that coincide with weaning. These biomechanical factors emanate from the contraction of temporalis and masseter and load the sagittal suture in tension to provide the mechanical strain sufficient to continue growth and resorption (i.e., remodeling). Staying congruent with the

560

BYRON

Fig. 5. Photographs of mouse calvariae taken with a dissecting microscope. Notice other calvarial sutures appear to increase in complexity in addition to the sagittal suture.

Fig. 6. Line plot showing skull width increasing with age. The rate of this increase is such that roughly 80% of total transverse cranial growth occurred by 10 days of age.

functional matrix paradigm, both brain growth and masticatory loading are considered covariates of cranial suture growth. Figure 7 demonstrates the correlation of suture complexity and skull width. The growth response from mastication is hypothetically observed by the increases in suture complexity long after transverse brain growth has plateaued (Fig. 4 vs. Fig. 6). This author posits mechanical forces resulting from mastication to play a more pivotal role in patterning cranial sutures than encephalization, as previously suggested (Behrents et al., 1978; Steenvoorden et al., 1990; Katsaros et al., 1994; Moss, 1997a; Kobayashi et al., 1999; Mommaerts et al., 2001; Kopher and Mao, 2003; Byron et al., 2004). This is supported by previous research by this author and col-

leagues (Byron et al., 2004) that demonstrate significantly increased bone formation rates and suture complexity in knockout mice with enhanced muscle function. This mechanobiological paradigm for suture patterning includes mechanoresponsiveness on the part of osteoblast (bone-forming) and osteoclast (bone-resorbing) cells. In the case of osteoclasts, it is unclear exactly how they “sense” their mechanical environment and help carry out the remodeling response. Similar skeletal connective tissue systems such as the periodontal ligament are known to respond to tension with bone formation and compression with bone resorption (Roberts-Harry and Sandy, 2004). Yu et al. (2004) utilized two-dimensional finiteelement analysis (FEA) in order to model a suture wave-

ONTOGENETIC CHANGES IN SUTURE OSTEOCLASTS

Fig. 7.

561

Skull width strongly correlates with, and thus can be considered a covariant of, suture complexity.

Fig. 8. Sagittal suture complexity correlates with osteoclast count when considering both alendronate and control treatment groups (n ⫽ 20).

form loaded in tension. It is felt that this is mechanobiologically relevant since several authors have observed tension as the primary loading regime of the sagittal suture upon contraction of the temporalis muscle (Behrents et al., 1978; Herring and Teng, 2000; Sun et al., 2004). Using 2D FEA, concave suture regions were predicted to experience increases in shear stresses while convex regions experience increases in tensile stresses. It is possible that shear is a signal that localized preosteoclasts are able to recognize in order to activate appropriately. However, this prediction fails to account for how osteoclasts robustly activate at early ages when the suture line is relatively straight and shear stresses are significantly lower. Clearly, there are several predictions that require further

testing and that shear along concave surfaces cannot account for the ontogenetic results observed above. Understanding the growth patterning of cranial sutures from a linear to a waveform shape is far from complete. The results presented here are interpreted as evidence that osteoclasts participate to model suture waveform and possibly receive guiding mechanobiologic signals. It is likely that early prenatal suture maintenance occurs through different pathways than postnatal suture maintenance. Kim et al. (1998) observed the expression of an important molecular craniofacial midline organizer, sonic hedgehog (Shh), and its receptor in a patched pattern along the osteogenic fronts of the embryonic sagittal suture. This suggests that an autonomous signal during

562

BYRON

Fig. 9. Representative TRAP-stained micrographs taken with the 20⫻ objective from the alendronate (A) and control (B) treatment groups. Note the discrete difference in TRAP enzyme positivity along the suture margins, especially along concave surfaces. This enzyme is labeled red by the TRAP stain and is noticeably absent from the alendronate-treated section.

prenatal development may relate to the subsequent serrated morphology seen in cranial sutures. Once this initial waveform is patterned, this author proposes a separate biomechanical suture maintenance pathway. The discovery that 10% equibiaxial strains are anabolic to calvarial osteoblasts is an important contribution (Fong et al., 2003a). In vivo confirmation of large strain regimes such as this in suture osteogenic tissues is necessary in a diverse sample of vertebrate organisms. Ideally, this sample would include small mammals such as rodents in addition to animals that are several orders of magnitude larger with respect to body size. By quantifying actual stress and strain levels during biting and feeding, we can begin to appreciate more completely this functional pathway for postnatal suture maintenance.

ACKNOWLEDGMENTS For input, suggestions, and analysis, the author thanks Mark W. Hamrick, Jack Yu, and Jim Borke. Cathy Pennington was also helpful in carrying out the protocols involved. Finally, the author thanks two anonymous reviewers for insightful and extremely helpful criticism.

LITERATURE CITED Baron R. 1999. Anatomy and ultrastructure of bone. In: Favus MJ, editor. Primer on the metabolic bone diseases and disorders of mineral metabolism, 4th ed. Philadelphia, PA: Lippincott Williams and Wilkins. p 3–10. Behrents RG, Carlson DS, Abdelnour T. 1978. In vivo analysis of bone strain about the sagittal suture in macaca mulatta during masticatory movements. J Dent Res 57:904 –908. Burke MJ, Winston KR, Williams S. 1995. Normal sutural fusion and the etiology of single sutural craniosynostosis: the microspicule hypothesis. Pediatr Neurosurg 22:241–246. Byron CD, Borke J, Yu J, Pashley D, Wingard CJ, Hamrick M. 2004. Effects of increased muscle mass on mouse sagittal suture morphology and mechanics. Anat Rec 279A:676 – 684. Carano A, Teitelbaum SL, Konsek JD, Schlesinger PH, Blair HC. 1990. Bisphosphonates directly inhibit the bone resorption activity of isolated avian osteoclasts in vitro. J Clin Invest 85:456 – 461.

Cohen MM Jr. 1997. Transforming growth factor beta s and fibroblast growth factors and their receptors: role in sutural biology and craniosynostosis. J Bone Miner Res 12:322–331. Cross SS. 1997. Fractals in pathology. J Pathol 182:1– 8. Enlow DH. 1990. Facial growth, 3rd ed. Philadelphia, PA: W.B. Saunders. Evans KD, Lau ST, Oberbauer AM, Martin RB. 2003. Alendronate affects long bone length and growth plate morphology in the oim mouse model for osteogenesis imperfecta. Bone 32:268 –274. Fong KD, Nacamuli RP, Loboa EG, Henderson JH, Fang TD, Song HM, Cowan CM, Warren SM, Carter DR, Longaker MT. 2003a. Equibiaxial tensile strain affects calvarial osteoblast biology. J Craniofac Surg 14:348 –355. Fong KD, Warren SM, Loboa EG, Henderson JH, Fang TD, Cowan CM, Carter DR, Longaker MT. 2003b. Mechanical strain affects dura mater biological processes: implications for immature calvarial healing. Plast Reconstr Surg 112:1312–1327. Green JR, Clezardin P. 2002. Mechanisms of bisphosphonate effects on osteoclasts, tumor cell growth, and metastasis. Am J Clin Oncol 25:S3–S9. Greenwald JA, Mehrara BJ, Spector JA, Warren SM, Crisera FE, Fagenholz PJ, Bouletreau PJ, Longaker MT. 2000. Regional differentiation of cranial suture-associated dura mater in vivo and in vitro: implications for suture fusion and patency. J Bone Miner Res 15:2413–2430. Hartwig WC. 1991. Fractal analysis of sagittal suture morphology. J Morphol 210:289 –298. Henderson JH, Longaker MT, Carter DR. 2004. Sutural bone deposition rate and strain magnitude during cranial development. Bone 34:271–280. Herring SW, Teng S. 2000. Strain in the braincase and its sutures during function. Am J Phys Anthropol 112:575–593. Jaslow CR. 1990. Mechanical properties of cranial sutures. J Biomech 23:313–321. Jaslow CR, Biewener AA. 1995. Strain patterns in the horncores, cranial bones and sutures of goats (Capra hircus) during impact loading. J Zool Lond 235:193–210. Kaku M, Kawata T, Kawasoko S, Fujita T, Tokimasa C, Tanne K. 1999. Remodeling of the sagittal suture in osteopetrotic (op/op) mice associated with cranial flat bone growth. J Craniofac Genet Dev Biol 19:109 –112.

ONTOGENETIC CHANGES IN SUTURE OSTEOCLASTS Katsaros C, Kiliaridis S, Berg R. 1994. Functional influence on sutural growth: a morphometric study in the anterior facial skeleton of the growing rat. Eur J Orthod 16:353–360. Kawata T, Tokimasa C, Fujita T, Kawasoko S, Kaku M, Sugiyama H, Tanne K. 1998. Midpalatal suture of osteopetrotic (op/op) mice exhibits immature fusion. Exp Anim 47:277–281. Kim HJ, Rice DPC, Kettunen PJ, Thesleff I. 1998. FGF-, BMP- and Shh-mediated signalling pathways in the regulation of cranial suture morphogenesis and calvarial bone development. Development 125:1241–1251. Kobayashi E, Hashimoto F, Kobayashi Y, Sakai E, Miyazaki Y, Kamiya T, Kobayashi K, Kato Y, Sakai H. 1999. Force-induced rapid changes in cell fate at midpalatal suture cartilage of growing rats. J Dent Res 78:1495–1504. Kopher RA, Mao JJ. 2003. Suture growth modulated by the oscillatory component of micromechanical strain. J Bone Miner Res 18:521– 528. Long CA, Long JE. 1992. Fractal dimensions of cranial sutures and waveforms. Acta Anat (Basel) 145:201–206. Lynnerup N, Jacobsen JC. 2003. Brief communication: age and fractal dimensions of human sagittal and coronal sutures. Am J Phys Anthropol 121:332–336. Ma YL, Bryant HU, Zeng Q, Schmidt A, Hoover J, Cole HW, Yao W, Jee WS, Sato M. 2003. New bone formation with teriparatide [human parathyroid hormone-(1-34)] is not retarded by long-term pretreatment with alendronate, estrogen, or raloxifene in ovariectomized rats. Endocrinology 144:2008 –2015. Mandlebrot B. 1967. How long is the coast of Britain? statistical self-similarity and fractional dimension. Science 156:636 – 638. Mandlebrot B. 1977. Fractals: form, chance, and dimension. San Francisco, CA: W.H. Freeman. Mao JJ. 2002. Mechanobiology of craniofacial sutures. J Dent Res 81:810 – 816. Mao JJ. 2005. Calvarial development: cells and mechanics. Curr Opin Orthop 16:331–337. Massler M, Schour I. 1951. The growth pattern of the cranial vault in the albino rat as measured by vital staining with alizarine red “S.” Anat Rec 110:83–101. Mommaerts MY, Staels PF, Casselman JW. 2001. The faith of a coronal suture grafted onto midline synostosis inducing dura and deprived from tensile stress. Cleft Palate Craniofac J 38:533–537. Monteiro LR, Lessa LG. 2000. Comparative analysis of cranial suture complexity in the genus Caiman (Crocodylia, Alligatoridae). Braz J Biol 60:689 – 694. Moss ML. 1957. Experimental alteration of sutural area morphology. Anat Rec 127:569 –590. Moss ML. 1961. Extrinsic determination of sutural area morphology in the rat calvaria. Acta Anat (Basel) 44:263–272. Moss ML. 1997a. The functional matrix hypothesis revisited: I, the role of mechanotransduction. Am J Orthod Dentofac Orthop 112:8 – 11. Moss ML. 1997b. The functional matrix hypothesis revisited: II, the role of an osseous connected cellular network. Am J Orthod Dentofac Orthop 112:221–226.

563

Moss ML. 1997c. The functional matrix hypothesis revisited: III, the genomic thesis. Am J Orthod Dentofac Orthop 112:338 –342. Moss ML. 1997d. The functional matrix hypothesis revisited: IV, the epigenetic antithesis and the resolving synthesis. Am J Orthod Dentofac Orthop 112:410 – 417. Opperman LA, Sweeney TM, Redmon J, Persing JA, Ogle RC. 1993. Tissue interactions with underlying dura mater inhibit osseous obliteration of developing cranial sutures. Dev Dyn 198:312–322. Opperman LA, Passarelli RW, Morgan EP, Reintjes M, Ogle RC. 1995. Cranial sutures require tissue interactions with dura mater to resist osseous obliteration in vitro. J Bone Miner Res 10:1978 – 1987. Opperman LA. 2000. Cranial sutures as intramembranous bone growth sites. Dev Dyn 219:472– 485. Rafferty KL, Herring SW. 1999. Craniofacial sutures: morphology, growth, and in vivo masticatory strains. J Morphol 242:167–179. Rice DP, Kim HJ, Thesleff I. 1997. Detection of gelatinase B expression reveals osteoclastic bone resorption as a feature of early calvarial bone development. Bone 21:479 – 486. Rice DP, Kim HJ, Thesleff I. 1999. Apoptosis in murine calvarial bone and suture development. Eur J Oral Sci 107:265–275. Roberts-Harry D, Sandy J. 2004. Orthodontics: 11, orthodontic tooth movement. Br Dent J 196:391–394. Steenvoorden GP, van de Velde JP, Prahl-Andersen B. 1990. The effect of duration and magnitude of tensile mechanical forces on sutural tissue in vivo. Eur J Orthod 12:330 –339. Sun Z, Lee E, Herring SW. 2004. Cranial sutures and bones: growth and fusion in relation to masticatory strain. Anat Rec (New Anat) 276A:150 –161. van Beek ER, Lowik CW, Papapoulos SE. 1997. Effect of alendronate treatment on the osteoclastogenic potential of bone marrow cells in mice. Bone 20:335–340. Vastardis H, Mulliken JB, Glowacki J. 2004. Unilateral coronal synostosis: a histomorphometric study. Cleft Palate Craniofac J 41:439 – 446. Wassersug RJ. 1976. A procedure for differential staining of cartilage and bone in whole formalin-fixed vertebrates. Stain Technol 51: 131–134. Winograd JM, Im MJ, Vander Kolk CA. 1997. Osteoblastic and osteoclastic activation in coronal sutures undergoing fusion ex vivo. Plast Reconstr Surg 100:1103–1112. Winograd JM, Im MJ, Vander Kolk CA. 2001. Enzymatic activiation associated with programmed fusion of the posterior interfrontal sutures in rats. Plast Reconstr Surg 108:927–937. Wolf G, Koskinen-Moffett L, Kokich V. 1985a. Migration of craniofacial periosteum in growing guinea-pigs. J Anat 140(Pt 2):245–258. Wolf G, Koskinen-Moffett L, Kokich V. 1985b. Migration of craniofacial periosteum in guinea-pigs with unilateral masticatory muscle paralysis. J Anat 140(Pt 2):259 –268. Yu JC, Borke JL, Zhang G. 2004. Brief synopsis of cranial sutures: optimization by adaptation. Semin Pediatr Neurol 11:249 –255.