High Impact Exercise Improves Bone Microstructure and Strength in Growing Rats

Physical activity is beneficial for skeletal development. However, impact sports during adolescence, leading to bone growth retardation and/or bone quality improvement, remains unexplained. This study investigated the effects of in vivo low (LI), medium (MI), and high (HI) impact loadings applied during puberty on bone growth, morphometry and biomechanics using a rat model. 4-week old rats (n = 30) were divided into control, sham, LI, MI, and HI groups. The impact was applied on the right tibiae, 5 days/week for 8 weeks mimicking walking (450 µε), uphill running (850 µε) and jumping (1250 µε) conditions. Trabecular and cortical parameters were determined by micro-CT, bone growth rate by calcein labeling and toluidine blue staining followed by histomorphometry. Bio-mechanical properties were evaluated from bending tests. HI group reduced rat body weight and food consumption compared to shams. Bone growth rate also decreased in MI and HI groups despite developing thicker hypertrophic and proliferative zone heights. HI group showed significant increment in bone mineral density, trabecular thickness, cortical and total surface area. Ultimate load and stiffness were also increased in MI and HI groups. We conclude that impact loading during adolescence reduces bone growth moderately but improves bone quality and biomechanics at the end of the growing period.

Bone growth rate and tibial length. Both HI and MI groups exhibited a reduction in proximal tibial growth rate (Fig. 2B), resulting in 8.7% and 5.6% decrease for HI and MI groups, respectively, compared to shams (Fig. 2B). Relative gross tibial length (value for control rats minus rats from other groups) exhibited significant differences for HI and MI groups (Fig. 2C). No significant difference was observed between control and sham groups for both growth rates (Fig. 2B) and relative tibial lengths (Fig. 2C).
Growth plate histomorphometry. Hypertrophic (HZ) and proliferative (PZ) zone thicknesses (Fig. 3A), as well as the number of proliferative cells per column and hypertrophic cells height (Fig. 3B), have been evaluated for all the experimental groups. MI and HI group exhibited significantly thicker (13% and 17%, respectively) HZ thickness compared to the sham group ( Fig. 3C-I). Moreover, PZ thickness also increased in HI groups (12%) compared to shams (Fig. 3C-II). Hypertrophic cell heights were also increased by 12% in HI group compared to the shams (Fig. 3C-III). The number of proliferative chondrocytes per column was similar for all groups (Fig. 3C-IV). Control vs. sham groups showed no significant difference in growth plate histomorphometric parameters (Fig. 3C).
Trabecular and cortical bone architecture. The effects of impact loading exercise on bone microstructure were assessed by comparing the impact groups with the shams after 4 and 8 weeks of repeated loading regime. For trabecular bone, the HI group showed an increase in BV/TV and a decrease in Tb.Sp (Table 1) after 4 weeks of loading. For cortical bone, Ct.Ar was increased in HI group at this time point (Table 2). After 8 weeks of loading, a significant increase was found in BMD and Tb.Th for both HI and MI groups (Table 1). However, only HI group showed a significant increment in BV/TV and a significant decrement in Tb.Sp at this time point (Table 1). For the cortical bone architecture, both HI and MI exhibited an increased Ps.Pm and Ct.Ar and a High impact loading triggers decreased body weight coupled with a reduced caloric consumption. Body weight was maximum for the control group followed by the shams and other impact groups at the end of the loading period (Fig. 1A). The body weight in HI group was decreased significantly compared to shams after 8 weeks of loading regime (Fig. 1A). Interestingly, food consumption was simultaneously reduced for the same group (HI) at the end of the study (Fig. 1B). The food consumption is generally dependent on the energy expenditure and so is the change in body weight 31 . Our findings showed that the HI group has less body weight and reduced food intake despite receiving the maximum intensity of the exercise regime. Part of this weight loss could be ascribed to the decreased appetite of the impact groups 32 , which was evidenced by a significantly lower caloric intake for the HI group (Fig. 1B). Another intriguing fact was the transitory fatigue of the exercised rats. We observed reduced activity limited to less than seven minutes after the forced loading regime. The rats continued their regular cage activity shortly after this phenomenon. It is suggested that the forced exercise can influence the levels of stress hormones and behavior of the animals which may lead to a reduction in caloric intake in the rats 33,34 . Moreover, bone osteocytes were shown to be sensitive to short term high-impact dynamic loading conditions 35 . It has also been reported that body weight reduction can activate a sensor dependent on osteocytes, which eventually diminishes caloric intake in the rodents [35][36][37] . The observed reduced body weight and caloric intake for the high impact group could have resulted from this phenomenon.
Our findings are supportive of other studies on adult animal models. Reduced body weight has been reported in trained animals by Jones et al. 38 and Huang et al. 31 , after 15 and 8 weeks of exercise period in adult rats, respectively. Moreover, simultaneously reduced body weight and reduced caloric consumption have been reported for adult running rats by Crew et al. 39 , and treadmill exercised in post pubertal rats by Tisuji et al. 40 and by Pitts and Bull 32 .

Figure 2.
Bone growth rates (µm/day) and longitudinal tibial lengths (mm). (A) 2.5x magnified microscopic images of the tibial metaphysis labeled twice with calcein and representative images of tibiae for control, sham, LI, MI and HI groups (I-V). Bone growth (ΔX, μm) measured as the mean distance between the two calcein lines, which were modeled as splines and divided by the time interval (3 days) between the two applied injections. (B) Bone growth rates (μm/day) of rat proximal tibiae for control, sham, LI, MI and HI groups. (C) Relative (control minus individual group) gross tibial length (mm) of the tibiae. MI and HI groups exhibited approximately three and four times reduction in tibial length difference. N = 6 rats per group (mean value ± SD). *p < 0.05 and **p < 0.01: significant compared to shams.  Medium and high impact loadings decrease longitudinal bone growth despite developing thicker HZ and PZ heights. Both MI and HI groups showed reduced bone growth rates at the proximal metaphysis compared to shams after 8 weeks of loading (Fig. 2B). This phenomenon eventually resulted in significant longitudinal growth retardation for the same two groups (Fig. 2C); it contradicts our hypothesis that longitudinal bone growth rate would remain unaffected under the impact loadings. Some noticeable histomorphometric changes were also concomitant along with this growth retardation. These changes include increased hypertrophic and proliferative zone thicknesses and hypertrophic cell heights (Fig. 3C).  www.nature.com/scientificreports www.nature.com/scientificreports/ The relationship between applied compression and longitudinal bone growth rate proposed by Hueter-Volkmann law states that increased compression reduces bone growth rate whereas reduced compression increases it 41,42 . Moreover, large compressive loads can lead to retardation of bone growth or even cease completely the bone growth [42][43][44][45] . Our findings are also consistent with other studies [46][47][48] , where rat ulna longitudinal growth was decreased by compressive loading in adolescence.

weeks of loading
Bone growth rate is generally correlated to the overall growth plate thickness 7,49 . Moreover, it is considered to be linearly correlated with the PZ 7,31 and HZ 27,50 thickness. Hence, the thicker HZ and PZ heights of the MI and HI group were expected to result in elongated bone length. Conversely, bone growth in MI and HI group was depressed even after thickening of growth plates. Previous studies have also reported thicker growth plates under excessive loadings. However, these studies related this phenomenon with dyschondroplasia (osteochondrosis) 46,51 . Osteochondroses are considered to be disorders of primary and secondary growth centers, or lesions at the apophyseal or epiphyseal growth areas of bones 52 . Active young athletes are prone to osteochondroses 52 , although it is not considered to be a permanent disability for diagnosed patients 53 . In most cases, conservative treatment for such symptoms includes sufficiently long rest 52 . In one form of such condition known as Scheuermann's disease 54 , regular physical exercises are even recommended. In our study, the rats have been sacrificed immediately after the repeated loading regime. So, it remains unclear whether a sufficient amount of unloading period would affect the growth plate thickness and change it accordingly or not.
The change in longitudinal bone growth could also be associated with the caloric intake and body weight of the rodents. It has been reported that the reduction of body weight due to reduced caloric intake can affect cell production in the proliferative zone in a negative manner 55 , which can eventually slow down longitudinal bone growth. Our observations can be compared with the studies using rat 56,57 and swine 55 models, where a reduction in body weight coupled with reduction in longitudinal bone growth has been observed. Our overall findings indicate that the generalized claim of the linear relationship between bone growth and growth plate height 7,58 might not always be implemented. However, our findings are supportive of other studies 7,46,47 , where a contrary relationship between the growth rate and height of the growth plate was also observed.

Changes in trabecular bone microstructure are time as well as impact level dependent.
Compared to the MI group, the HI group showed load adaptive changes in trabecular microstructure at an earlier (after 4 weeks) period and affecting more structural parameters (Table 1). BMD was significantly increased in both MI and HI groups (Table 1) at the end of loading period. For healthy bone structure, bone mineral content shows an increasing trend during the adolescent period 59 . Also, higher loading intensity is generally associated with an increased BMD 60 . In fact, BMD in athletes is elevated under high impact training conditions 61 . The increase of BMD in our study could be related to hormones triggering mechanisms. Indeed, an increased BMD is controlled by a decreased parathyroid hormone response coupled with an increased calcitonin response 62,63 , both of which take time to react under favorable loading conditions. Hence, this could explain the significant increment of BMD assessed after 8 weeks instead of 4 weeks of loading. In similar studies, where adult rats have been tested for repetitive jumping exercise 64 or treadmill running exercise 65 , an increased BMD was observed in the loaded limbs.
The BV/TV was also found to significantly increase in the exercised tibiae both after 4 and 8 weeks of loading in the HI group. For a healthy growing bone, an elevated BV/TV is generally correlated with an increased BMD 66 , as found in this study. The significant BV/TV in HI group after 8 weeks of loading can be explained with the increased BMD for the same group (Table 1). However, the significant increase after 4 weeks of exercise (without simultaneous BMD increase) could be related to triggering of bone metabolism under high impact loading 61 . Indeed, it has been reported that under controlled loading scenarios, the trabecular structure responds positively through diffusion and active transport of metabolites within the entire microstructure 59,67 . So, it could be possible that HI loading has accelerated the diffusion and transportation process of the metabolites at an earlier stage and thus elevated the BV/TV in the exercised limb accordingly. Other studies support our findings in growing rats, where swimming exercise was found to significantly increase BV/TV 68,69 . Also, another study reported a greater BV/TV in the growing rat tibiae after 8-weeks of free fall exercise routine 70 . The observed increment in Tb.Th during the loading period is a part of normal bone development 71 . However, a significant increment in the loaded limbs (compared to shams) indicated an additional improvement in trabecular structure under impact loading conditions. The significant reduction in Tb.Sp is often considered as the concomitant increase with BV/TV and Tb.Th 72 . The significant reduction in Tb.Sp indicates the occurrence of a loading induced bone gain through increased connectivity and gradual thickening of the trabecular structure 73 . The significant change in Tb.Sp in HI group at an earlier stage can be associated with the significant increase in BV/TV from the same group (Table 1). Our data are supportive of previous findings where an increased Tb.Th was reported in the loaded tibiae of 10-week old adult mice 74 , as well as with a decreased Tb.Sp reported in loaded tibiae of both growing and adult mice 15 . Medium and high impact loadings benefit the cortical bone morphometry in the diaphysis, leading to significantly improved structural-and tissue-level bending mechanical properties. MI and HI loadings significantly affected tibial diaphysis, modifying both its cortical microstructure and its mechanical properties (Tables 2 and 3); it supports the hypothesis that bone morphometry and biomechanics are improved by impact loadings and that higher impact intensity has greater positive influence on bone morphometry and biomechanics. Indeed, stiffness was significantly increased in these groups compared to the sham group (Table 3). This improved stiffness eventually triggered the bones to reach a significantly higher ultimate load (Table 3). Interestingly, MI and HI groups exhibited significantly increased cortical area, simultaneously coupled with periosteal perimeter expansion and endocortical perimeter reduction ( www.nature.com/scientificreports www.nature.com/scientificreports/ the mid-diaphysis and the corresponding ultimate load (F ult ) are highly correlated to each other 75,76 . Hence, the increased ultimate loads for MI and HI groups can be justified from their significantly increased cortical area. Our findings agree with other studies on adult rodents, where an increased ultimate strength have been associated with exercised tibiae in 15 and 10 week old rats 31 and mice 74 , respectively. Bone stiffness can be related to its morphology and cross-sectional geometry in growing rats 77 . More specifically, stiffness can be associated with the increase or decrease in cortical thickness (Ct.Th), total area (Tt.Ar) and cortical area (Ct.Ar) 78,79 . Ct.Ar has been increased significantly for both MI and HI groups (Table 2). Also, Ct.Th and Tt.Ar significantly increased for the HI group (Table 2). Hence, a strengthened diaphysis associated with an increased stiffness can be justifiable for the MI and HI groups. Increased ultimate load and stiffness was also observed in a study 31 of 15-week old swimming and running rat groups along with an increased cortical area in the swimming groups. Another study with 10 week old mice reported an increased cortical thickness and stiffness, along with the cortical and total area increment in the loaded limbs 74 .
No effects were found on Young's modulus when comparing LI, MI and HI loadings to shams (Table 3). Young's modulus or bone rigidity is an intrinsic mechanical property of the bone 80,81 . With the greater structural strength found in the MI and HI groups (Table 3), an increase was also expected in the Young's modulus. In a bending test, the evaluation of rigidity depends linearly on stiffness and cubicly on the span length of the tibiae (Equation 2) 82 . In the MI and HI groups, tibial lengths decreased by 4% and 5%, respectively. Even though their stiffnesses increased, the decreased span lengths in the cubic form might have counteracted this change, yielding unchanged Young's moduli. Another study also reported non-significant changes in Young's modulus in growing mice limbs 74 undergoing 2 weeks of axial loading regime generating a maximum 2400 με at the mid-diaphysis of the tibiae.
Energy to failure load increased significantly for the MI and HI groups ( Table 3). Greater energy to failure loads implies that the MI and HI tibiae sustained more deformation or strain before failure. Strengthening of bone tissue in the MI and HI groups is associated with its adaptation in response to the applied loading regime. During the loading period, compositional alterations might have occurred within the newly forming and pre-existing bone cellular matrix 83 , possibly involving type I collagen, which is known to affect the post-yield behavior of the bone 84,85 . Biochemical changes might have also interfered with collagen fibers orientation within the bone matrix and thus altered the toughening-mechanism in the bone microstructure 86 . Ultimate stress (σ ult ) and failure stress (σ fail ) are significantly increased and decreased respectively in both MI and HI groups ( Table 3). The ultimate/failure load is the controlling factor in the corresponding ultimate/failure stress for the same group of tibiae (Equation 3) 31,87 . The respective significant increase and decrease in F ult and F fail for the MI and HI groups can explain the significant change in σ ult and σ fail for the same groups. Our results agree with other studies, where 15-week old rats exhibited increased stress and failure loads in the trained limbs 31,88 . Moreover, studies using 8-week old mice also observed increased ultimate stress in the loaded limbs after 2 weeks of axial loading regime 74 .
Limitations. The current study has limitations involving some methodological aspects. For the trabecular VOI, only the proximal metaphysis was analyzed. It was chosen over the distal section because of the large amount of trabeculae in this region. Also, it has been reported that the proximal tibia has greater bone volume compared to the distal tibia 89 and has also been used more often in bone remodeling studies 16,19,74 . As for bone growth rate, it was only measured at the proximal site. This choice was justified as proximal metaphysis is responsible for blood supply and vascular stasis in growing bone and the contribution of the proximal tibial growth plate in the total longitudinal growth was found to reach approximately 80% in adolescents 90 . So, any loading effects on proximal bony region could presumably be considered as to have the main effects between the proximal and distal parts as well. Also, the biological effects of bone remodeling were not investigated as part of this study. A future study could investigate bio-markers to infer on bone formation and resorption for better understanding of bone growth mechanism involved in impact loading regimes. Moreover, rats were sacrificed immediately after the last in vivo loading regime at 81 days old. A detraining period before the sacrifice, which might have modified the growth plate histomorphometry 47,91 , was not evaluated as our primary objective was to investigate the bone growth rate and histomorphometry while the growth plate was still active 92,93 .

Conclusion
A significant decrease in body weight associated with a reduction in food intake was found for the high impact group. A thicker growth plate was observed for medium and high impact group despite having a significant longitudinal growth retardation. Also, the medium and high impact groups benefitted the trabecular and cortical microstructure and led to significant changes in structural-and tissue-level mechanical behavior in the diaphysis. The low impact group also altered bone structure, by exhibiting an increasing or decreasing trend in certain bone microstructural properties, but these changes were not statistically significant. In summary, a brief (10 min) daily exercise of medium (850 µε) and high (1250 µε) impact physical activity during adolescence can influence bone growth, and improve the quality and mechanics of bone microstructure at the end of the growing period. This loading model provides the scope to fully understand the role of controlled mechanical loading during the adolescent period and will be used in future for the design of non-invasive loading models to modify the bone microstructure and mechanical strength for building up a healthy and robust skeleton. Future work will investigate whether these impact loads applied during puberty are determinant factors for bone quality and strength in adult life.

Materials and Methods
Animals. Animals (n = 30, male, Sprague-Dawley rats) were obtained from Charles River Laboratories www.nature.com/scientificreports www.nature.com/scientificreports/ University Hospital, Montreal, Canada. Rats were housed two per cage at 25 °C with a 12-h light/dark cycle. Standard laboratory diet and water ad libitum were provided. Three days prior to the beginning of loading, rats were randomly divided into five groups (n = 6/group): control, sham, low impact (LI), medium impact (MI), and high impact (HI). Body weights and food intakes were monitored weekly from the beginning (4 th week) of in vivo loading until the end (11 th week) to monitor overall health.
In vivo axial tibial loading. While the rats of the LI, MI, and HI groups were kept anesthetized (2% isoflurane, 1.0 L/min O 2 ), the cyclic impact loading was applied to the right tibia with a custom-built impact loading device (Fig. 4A). The device was controlled using a Mach-1 V500C (Biomomentum Inc., Montreal, Canada). Haversine waveform displacements 94 (derived from the calibration experiment) were applied at 2 Hz and characterized by symmetric active loading/unloading with a 0.10 sec of rest insertion between load cycles 48,95 (Fig. 4B). The frequency of 2 Hz was selected as it matches with the stride frequency range observed in rat normal locomotion 48 . A compressive preload of 0.5 N was applied to keep the tibia in position. Loadings were repeated for 1200 cycles, yielding a daily (5 days/week) loading period of 10 min 16 (Fig. 4B). Sham rats received the same experimental setup conditions without any load application. Controls were kept in the cages without any manipulation. Regular cage activity was allowed between loading sessions.
Strain calibrated impact loading. Prior to implementing the loading experiments, impact parameters were strain calibrated. The relationship between applied displacement and bone tissue deformation for the right tibia was established in an in vivo compression-strain calibration experiment 94 for the 4, 8 and 12-week old rats. This relationship was adapted for determining the required displacement to generate 450, 850, and 1250 µε at the medio proximal surface of the rat tibia [96][97][98][99] . These strain magnitudes were chosen to correspond to peak strain values in the human tibia during unrestricted walking (450 µε), zig-zag uphill running (850 µε), and vertical jumping (1250 µε) conditions [96][97][98][99] . Animals (n = 18, male, Sprague-Dawley rats) were obtained from Charles River Laboratories (Montreal, Canada) at approximately 3, 7 and 11 weeks of age (n = 6/age group) and allowed to acclimate for one week before the strain calibration experiment. After CO 2 asphyxiation, followed by decapitation, www.nature.com/scientificreports www.nature.com/scientificreports/ right tibiae were collected for each age group of rats. Tibiae were isolated from the middle of the femur to the toes. After tibial collection, an incision was made near the medio proximal surface of the tibia. Overlying skin and muscles were retracted to expose the bonding surface, polished with an abrasive paper and cleaned with ethanol solution. A single element strain gauge (C2A-06-015LW-120; dimensions: 0.86 mm × 1.32 mm, Micro-Measurements Group, Raleigh, NC, USA) was bonded with cyanoacrylate (M-Bond 200; Micro-Measurements Group) at 35% of the tibial length (L) (Fig. 4C). A compressive preload of 0.5 N was applied to keep the tibia in position (Fig. 4C). Haversine waveform displacements were applied at 2 Hz with a 0.10-sec rest insertion between displacement cycles. Displacements ranging from 0.5 mm to 3.5 mm were applied with a 0.50 mm increment. The strain data were recorded simultaneously at 2.5 kHz with a PC via Labview software (Labview 8.6, NI) through the quarter-bridge completion and analog input modules. Applied displacement and resulting strain were also recorded simultaneously. Displacement versus strain curves was plotted for three age groups, and a linear fit was applied to obtain the compression-strain relationships (Fig. 4D). Using these calibration curves, the amount of axial displacements were determined for producing the target tensile strain (450, 850, and 1250 µε) in LI, MI, and HI group of rat tibiae, respectively. The axial displacement values for the week's in-between the chosen calibration studies were adjusted by linear interpolation using the two known displacement values and the weekly-age of rats.
Micro computed tomography (micro-CT). Weekly scanning regime. A micro-CT scanner was used to perform eight weekly CT scans of the right tibia of the rats, from 4 to 11 weeks of age. The imaging system was a Skyscan 1176 in-vivo micro-CT (Skyscan, N.V., Belgium) scanner with rotatable X-ray source and detector. Following the weekly loading regime, rats were anesthetized (2% isoflurane, 1.0 L/min O 2 ), positioned on the carbon fiber half-tube bed of the Skyscan 1176, and maintained on anesthetic gasses for the duration of the scanning process (5-12 min). A cylindrical shape Styrofoam holder was used to position the right tibia to ensure the placement of the tibia along the scanner midline. All scans were performed with 18 μm voxel resolution, 65 kV, 384 μA, 350 ms exposure time, 0.65° rotation step, and a 1-mm Al filter 93 . A Phantom calibration was performed for each scanning day, prior to bone imaging using two cylindrical hydroxyapatite phantoms (0.25 and 0.75 g/cm 3 CaHA). Images were reconstructed using NRecon software (v.1.6.10, Bruker-microCT, Belgium) 93 .
Trabecular bone morphometry. The volume of interest (VOI) for trabecular bone was selected as a percentage of the entire tibial length (L) (Fig. 5A). The VOI started at ~0.35 mm distal to the growth plate, excluding the primary spongiosa, and extended for 12% of the total tibial length (L) 19,93 (Fig. 5A). The volumes of interest, including only trabecular bone, were semi-automatically segmented using an in-house algorithm 93 . For all analyses, a global gray threshold value of 65 corresponding to an equivalent density of 0.413 g/cm 3 of calcium hydroxyapatite (CaHA) was applied 19,93 . CTAn software v.1.16 was used for performing morphometric analysis for the selected VOIs of trabecular bone and allowed evaluating the following parameters: bone mineral density (BMD), bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N), trabecular spacing (Tb.Sp), and connectivity density (Conn.Dn) 100 (Fig. 5A).
Cortical bone morphometry. The VOI for cortical bone was chosen from the tibial mid-diaphysis and extended proximal-distally for a total of 5% of the tibial length (Fig. 5A). The volumes of interest, including only cortical . The trabecular VOI started at ~0.35 mm distal to the growth plate and extended for 12% of the overall bone length (L). The VOI for cortical bone was centered at the tibial mid-diaphysis and extended proximally and distally for 5% of the tibial length (L). Volumes of interest including only trabecular and cortical bone were semi-automatically segmented using an in-house algorithm. (B) I-Experimental setup for the three-point bending tests. A distance of 50% of the total tibial length was fixed between the supports, while the remaining 50% was distributed equally between the external sides of the supports. II-Representative image of the fractured tibia after the bending test. www.nature.com/scientificreports www.nature.com/scientificreports/ bone, were semi-automatically segmented using an in-house algorithm 93 . A global gray threshold value of 65 was set for all analyses 19,93 . Cortical microarchitectural measurements, including tissue mineral density (TMD), cortical bone area (Ct.Ar), total area (Tt.Ar), medullary area (Ma.Ar), cortical thickness (Ct.Th), endocortical perimeter (Ec.Pm), periosteum perimeter (Ps.Pm), and mean eccentricity (Ecc) were evaluated from the cortical bone VOIs 100 (Fig. 5A).
Mechanical testing. Following the last micro-CT scan (11 th week), rats were sacrificed using CO 2 asphyxiation, followed by decapitation. Right tibiae (n = 30) from all groups of rats were cleaned of soft tissues and tested to failure in three-point bending under displacement control at 0.20 mm/sec using a Mach-1 V500C (Biomomentum Inc, Montreal, Canada). A custom made bending setup was used to place the rat tibiae horizontally with the anterior surface upwards and centered between the supports (Fig. 5B). A distance of 50% of the total tibial length was set between the supports, while the remaining 50% was distributed equally between the external sides of the supports (Fig. 5B). The structural properties of the tibial samples, including the ultimate load (N), failure load (N), energy to ultimate load (mJ), energy to failure load (mJ), and linear stiffness (N/mm) were determined using force-deformation curves. Energies to ultimate load and failure load were computed as the areas under the force-deformation curves. Stiffness was calculated as the slope of the linear portion of the force-deformation curve. To calculate the intrinsic cortical mechanical properties, cross-sectional parameters were measured using the micro-CT images at the tibial mid-diaphysis 87 . Assuming the tibial cross sections were elliptically shaped 101 , the moment of inertia was evaluated using the following equation: (1) 3 3 where I is the moment of inertia (mm 4 ), a is the width of the bone cross-section in the mediolateral direction (mm), b is the width of the bone cross-section in the anteroposterior direction (mm), and t is the average of the cortical thickness (mm) 31 . Assuming linear elastic bone material 82,102,103 and using the beam bending theory, Young's modulus and ultimate stress were determined by the following equations: where, E is the Young's modulus (GPa), and σ ult is the ultimate stress (MPa), k is the stiffness (N.mm −1 ), L is the span length (mm), F ult is the ultimate load (N), and c is the distance from the cross-section centroid to outermost point on the cross-section (mm), which was approximated as bone width/2 87 . Moreover, energies to ultimate stress (mJ/mm 3 ) and failure stress (mJ/mm 3 ) were assessed by calculating the respective areas under the stress-strain curve.
Bone growth rate assessment. Calcein injections. Calcein was used to mark the newly formed bone line on the proximal tibial surface to enable longitudinal bone growth rate measurement. Injections of calcein (Sigma-Aldrich, St. Louis, MO, USA) were made intraperitoneally at a dosage of 15 mg/Kg 44 at 5 and 2 days prior to euthanasia.
Tissue processing. Proximal sections (~10 mm) from each tibia were fixed for 48 h using the formalin solution (Anachemia, Montreal, QC, Canada). For dehydration and clarification, graded alcohol solutions and xylene were used respectively. Embedding process was performed using methylmethacrylate (MMA) (Fisher Scientific Canada, Nepean, ON, Canada) 43 . After polymerization, tibial blocks were cut in 6 µm sections using a microtome (Leica SM2500). The longitudinal bone axis was used to cut the tibiae for 36 slides, six series of six slides, containing two sections per slide. For each tibia, the first slide of each series was set aside from light for growth rate measurements, for a total of 6 slides (i.e., 12 sections total). Microscopic observations were performed with a microscope (Leica DMR with Retina Qimaging Camera) using 2.5x magnification.
Growth rate calculation. Bone growth rate was calculated as the distance between two calcein labels divided by the time interval (3 days) between injections 104 ( Fig. 2A). Measurements were performed using a custom-made Matlab program, where both calcein lines were modeled as splines, and the distance between the labels was automatically calculated as the mean value of 100 segments parallel to the longitudinal growth direction 43,44 (Fig. 2A).
Growth plate histomorphometry. Growth plate histomorphometric parameters were assessed to infer on the effects of impact loadings, similar to previous studies on bone growth mechanobiology 42,43 . These parameters included the PZ and HZ heights, hypertrophic cell height, as well as the number of proliferative cells per column 43,44 (Fig. 3A,B). Zone heights and hypertrophic cell height were measured by implementing a 10x and 20x magnified image sets, respectively, following a similar approach to bone growth rate measurement. Average zonal height values were calculated from 100 individual segmental measurements (Fig. 3A). Also, 20x magnified image sets were used for measuring the number of proliferative chondrocytes per column (Fig. 3B). For each proximal tibial segment, histomorphometric parameters were measured by averaging 72 values, 6 values per section, and 12 values per microscope slide with a six series repetition.