Multiscale experimental study on the effects of different weight-bearing levels during moderate treadmill exercise on bone quality in growing female rats

Juan Fang; Jiazi Gao; He Gong; Tianlong Zhang; Rui Zhang; Bangchao Zhan

doi:10.1186/s12938-019-0654-1

. 2019 Mar 22;18:33. doi: 10.1186/s12938-019-0654-1

Multiscale experimental study on the effects of different weight-bearing levels during moderate treadmill exercise on bone quality in growing female rats

Juan Fang ^1,², Jiazi Gao ¹, He Gong ^1,^✉, Tianlong Zhang ¹, Rui Zhang ³, Bangchao Zhan ¹

PMCID: PMC6431042 PMID: 30902108

Abstract

Background

Bone tissue displays a hierarchical organization. Mechanical environments influence bone mass and structure. This study aimed to explore the effects of different mechanical stimuli on growing bone properties at macro–micro–nano scales.

Methods

Sixty five-week-old female Wistar rats were treadmill exercised at moderate intensity with the speed of 12 m/min, and then randomly divided into five groups according to weight-bearing level. After 8 weeks of experiment, femurs were harvested to perform multiscale tests.

Results

Bone formation was significantly increased by weight-bearing exercise, whereas bone resorption was not significantly inhibited. Trabecular and cortical bone mineral densities showed no significant increase by weight-bearing exercise. The microstructure of trabecular bone was significantly improved by 12% weight-bearing exercise. However, similar positive effects were not observed with further increase in weight-bearing levels. The nanomechanical properties of trabecular bone were not significantly changed by weight-bearing exercise. The macrostrength of whole femur and the nanomechanical properties of cortical bone significantly decreased in the 19% and 26% weight-bearing exercise groups.

Conclusion

When rats ran on the treadmill at moderate intensity during growth period, additional 12% weight-bearing level could significantly increase bone formation, improve microstructure of trabecular bone, as well as maintain the structure and mechanical properties of cortical bone. Excessive weight-bearing level caused no positive effects on the trabecular bone microstructure and properties of cortical bone at all scales. In addition, increased weight-bearing level exerted no significant influence on trabecular and cortical bone mineral densities.

Keywords: Weight-bearing, Moderate intensity, Treadmill exercise, Bone quality

Background

Running is one of the most common weight-bearing exercises and it can increase bone mass in humans and animals [1–5]. Numerous experiments show that the effect of treadmill exercise on bone mass is related to exercise intensity and the effect of the moderate-intensity treadmill exercise on increasing bone mass and strength is better than those of low- and high-intensity treadmill exercise [3–10]. Thus, the moderate-intensity treadmill exercise may be the most suitable form within these three exercise intensity levels.

The effect of treadmill exercise on bone mass is associated with the peak strain on the bone produced by exercise; and animal studies have reported that once the mechanical strain induced by exercise exceeds a certain threshold, bone formation is positively correlated with the peak strain magnitude [10–12]. Treadmill running may increase more bone mass in the long bones of rats at the main weight-bearing sites than the less weight-bearing sites [13]. Increasing the weight of backpack can effectively increase the peak strain, thus it can further effectively increase bone mass [14–19]. A previous study reports that moderate-intensity treadmill exercise with a 19–20% body weight backpack can significantly increase bone mineral content in the hind limbs of 5-month-old female rats [18]. This form of exercise also significantly increases the bone mineral density (BMD) and bone mineral content of trabecular bone in 3-month-old ovariectomized female rats [19]. Accordingly, moderate treadmill exercise in combination with a certain additional load in backpack may effectively increase bone mass in adult rats.

The responses of bone to mechanical stimuli are distinct at different growth stages, and the sensitivity of bone to mechanical loading is significantly higher in young mice than in adult and aged animals [20]. Treadmill exercise during the growth period has positive effective on bone quality [1, 14, 15, 21–25]. Most of the subjects in previous studies are adult rats, and only one weight-bearing level is established. The effect of treadmill exercise with different weight-bearing levels on bone quality in growing rats remains unclear. Our previous study investigated the effects of treadmill exercise with different weight-bearing levels on the subchondral trabecular bone microstructure and articular cartilage morphology of the knee joint in growing female rats [26]. Results showed that the effect of treadmill exercise on articular cartilage is associated with weight-bearing levels. The microstructure of subchondral trabecular bone is significantly improved by 12% weight-bearing level. Nevertheless, the bone mineral density (BMD) and bone volume fraction (BV/TV) are decreased, and the structure model index (SMI) is increased by 26% weight-bearing level; thus, this exercise exerts no positive effects on the microstructure of trabecular bone [26]. However, in our previous study, the influence of treadmill exercise with weight-bearing is only investigated on articular cartilage in growing female rats. The effect of this form of mechanical load on bone quality remains unknown.

Accordingly, moderate treadmill exercises with different weight-bearing levels groups were established in this study. Bone displays hierarchical structural organization and has different properties at macro–micro–nano scales [27]. Bone properties may be affected by treadmill exercise from macro to nano scale. The aim of this study was to explore the effect of different weight-bearing levels during moderate treadmill exercise on bone quality (i.e. BMD, bone geometry, microstructure, macromechanical and nanomechanical properties) in growing female rats at macro–micro–nano scales.

Methods

Experimental design

This study was performed in strict accordance with the recommendations of the Laboratory Animal Standardization Committee. The protocol was approved by the Medical Ethics Committee of the No. 1 Hospital of Jilin University (No. 2013-145).

A total of 60 5-week-old female Wistar rats were purchased from the Experimental Animal Center of Jilin University. All rats with the body weight 123.07 ± 16.33 g arrived at the laboratory a week before the experiment began to accommodate the diet and new environment. These rats were provided with standard rodent diet (autoclaved NIH-3 with 6% fat; 18% protein; Ca:P, 1:1; and fortified with vitamins and minerals) and tap water during the experimental period. The environmental temperature was 24 ± 2 °C under natural light conditions. All rats were randomly divided into five groups with 12 animals each. The schematic of experimental grouping is shown in Fig. 1a. Weight-bearing was in the form of backpack; the backpack weights were 5%, 12%, 19%, and 26% of the individual body weight of rats. The five groups are as follows: treadmill exercise control group (ex), treadmill exercise with 5% weight-bearing level group (eb5), treadmill exercise with 12% weight-bearing level group (eb12), treadmill exercise with 19% weight-bearing level group (eb19), and treadmill exercise with 26% weight-bearing level group (eb26).

Fig. 1 — The schematic of experimental grouping and the measurements performed in the femur. a Experimental grouping. b Measurement of the femur

Our previous study demonstrates that, according to blood lactate accumulation, for 5-week-old female rats, treadmill exercise at 12 m/min for 15 min correspond to moderate intensity exercise [26]. Therefore, all rats in this experiment ran at 12 m/min for 15 min in each day.

Exercise protocol

All rats exercised 7 days/week for 8 weeks. During the first 2 weeks, the rats were conditioned to treadmill exercise and wore backpack. In this period, velocity was gradually increased from 8 m/min to 12 m/min with the increment of 3% per day for 15 min/day and 7 days/week. The backpack weight of rats in all weight-bearing groups was gradually increased from 0 to the target weight, with the increment of 7% of the target weight borne by the rats in each group per day. The duration of treadmill exercise with weight-bearing groups was 15 min/day, and no weights at other time points. The form of backpack is consistent with that used in the literature, and the backpack is filled with leaden strips [18]. The body weight of each rats were measured once a week. The load in the backpack was adjusted to the change in body weight.

Specimens preparation

At 13 days before killing, all rats were treated with the subcutaneous injection of calcein (5 mg/kg). At 3 days before killing, calcein were re-injected to create a fluorescence marker [23, 28–30]. After 8 weeks of experiment, the rats were killed by collecting blood samples from the abdominal aorta under pentobarbital sodium anesthesia, and the left lower limb calf muscles (gastrocnemius and soleus muscles) of each animal were removed and weighed immediately. Bilateral femurs were prepared for testing, following removal of skin, muscles, and tendons. The length of left femurs was measured from femoral head to the distal medial condyle (Fig. 1b). The maximum and minimum external diameters positioned at 1/2 of the total left femur were measured (Fig. 1b), and the cross-sectional area was calculated. Schematic of the measurements performed in this study is shown in Fig. 1b.

Serum test

Not less than 5 ml of blood was collected from the abdominal aorta under general anesthesia and centrifuged at 3000 rpm/min for 15 min to obtain serum samples. The alkaline phosphatase (ALP) and tartrate-resistant acid phosphatase (TRACP) in the serum were marked by enzyme linked immunosorbent assay (ELISA kit, Nanjing, China) using a Microplate reader (ELx800, BioTek Corporation, USA). Afterward, the rates of bone formation and resorption were determined.

Three-point bending mechanical test

The right femurs were cleaned and kept wet during test by normal saline. Three-point bending mechanical test was conducted with the right femur loaded at mid-length in the anterior–posterior direction using an electronic universal testing machine (AG–X plus, Shimadzu, Japan). The test was continued until fracture, with a fulcrum span of 20 mm and an actuator velocity of 1 mm/min. Deflection (mm) and failure load (N) were obtained from load–displacement curve. Energy absorption (J) was defined by integrating the force–displacement curve from zero to the maximum force. Stiffness (N/mm) was determined from the slope of the linear portion (10–60% maximum force) of the load–displacement curve. Macro-elastic modulus (E) of the femur is calculated using the following equation:

E = F l^{3} / 48 I f

where F is failure load, l is span, f is deflection, and I is moment of inertia of fracture surface caused by three-point bending mechanical test, which is calculated from the micro-CT image.

Histomorphometry

After three-point bending mechanical test, the right proximal femurs were fixed with 70% concentration ethanol and dehydrated progressively in 70%, 80%, 90%, and 100% EtOH (24–48 h per phase), and all proximal femur samples were embedded separately in polymethylmethacrylate. The embedded samples were cut along the coronal plane with a low-velocity diamond saw under constant deionized water irrigation, from which the proximal trabecular bone was exposed. For histomorphometry, 5-mm thick slices were cut along the coronal plane with a thin-sliced cutting machine for hard materials.

Laser scanning confocal microscopy (FV500, Olympus Corporation, Japan) was used to obtain images of right proximal femur samples, and one image per right femur sample was obtained. The mineral apposition rate (MAR), and the ratio of mineralizing surface to bone surface (MS/BS) of trabecular bone of 1 mm² region of interest (ROI) in the right femoral head were calculated according to the standard nomenclature described by Parfitt et al. measured by Image-Pro Plus software [31, 32].

Micro-CT scanning

The microarchitecture of the left femur was scanned by a micro-CT system (Skyscan 1076, Skyscan, Belgium), and then quantitative analysis of the 3D microarchitectures of trabecular bone in the femoral head and cortical bone in the femoral shaft was performed. The spatial resolution for specimen scanning was 18 μm. The scanned data was reconstructed by using NRecon software (NRecon, Skyscan, Belgium). BV/TV, Tb.Th, Tb.N, Tb.Sp, and BMD of trabecular bone (Tb.BMD) in the femoral head, and BMD, and thickness of cortical bone (i.e. Ct.BMD, Ct.Th) in the femoral shaft were calculated by CTAn (CTAn, Skyscan, Belgium).

The ROI in the micro-CT images was selected manually. The ROI for femoral head was all trabecular bone within the femoral head, and trabecular bone was selected as much as possible in the ROI. The ROI for the femoral shaft was positioned at 1/2 of the total left femur with 2 mm-thick from distal to proximal (Fig. 1b) [33], and 106 2D micro-CT images of each sample were selected.

Nanoindentation test

After micro-CT scanning, longitudinal trabecular bone samples with a thickness of 2 mm cut from the left femoral heads along the axis of femoral neck (Fig. 1b) were used for nanoindentation test, and then the longitudinal indentation modulus (E_Tb.L) and longitudinal hardness (H_Tb.L) of trabecula were measured. Longitudinal cortical bone samples with a thickness of 2 mm cut from the femoral shaft along the axis of the femoral shaft were used for nanoindentation test, and then the longitudinal indentation modulus (E_Ct.L) and longitudinal hardness (H_Ct.L) of cortical bone were measured. The femoral shaft was cut along its axis to expose the transverse section of cortical bone, and transverse cortical bone samples were obtained for nanoindentation test; the transverse indentation modulus (E_Ct.T) and transverse hardness (H_Ct.T) of cortical bone were measured. All samples were dehydrated similar to the right femur, and the dehydrated samples were embedded in epoxy resin [34]. All embedded samples were metallographically polished using silicon carbide abrasive papers of decreasing grit sizes (400, 800, 1500, 2000, 3000, and 5000 grit), and finally on microcloths with finer grades of diamond suspensions to the finest, 0.05 μm grit, to produce smooth surfaces for nanoindentation test. The specimens were washed in deionized water between each polishing step to remove debris.

Nanoindentation tests were performed using Nano Indenter G200 (Nano G200, Agilent Technologies, Ltd., USA) with a Berkovich diamond indenter. The test was performed by driving the indenter into the specimen to a depth of 1000 nm at a constant loading rate of 750 μN/s (held for 10 s) and then unloading to 15% of the peak load at a rate equal to half that used during loading [23].

For trabecular bone samples, all indents were conducted at the similar site based on the optical microscopy observation to eliminate any local effects (Fig. 2), 5 indentations were made in each sample, and then the mean value was calculated, as well as each cortical specimen. The indentation modulus (E) and hardness (H) were determined using the method of Oliver and Pharr [35].

Fig. 2 — Indentation sites for nanoindentation test. Actual indented sites were marked by red cross under optical microscopy, and the sample thickness was 2 mm

H is calculated by the following equation:

H = P_{max} / A

where P_max is peak load, and A is contact projected area of the Berkovich indenter tip.

The relationship between effective modulus for the indenter-specimen combination (E_ef) and slope (S) of the linear stage in the upper part of the unloading curve is:

S = 2 π^{- 1 / 2} β E_{ef} A^{1 / 2}

Effective modulus for the indenter-specimen combination E_ef can be derived from:

1 / E_{ef} = (1 - ν_{b}^{2}) / E_{b} + (1 - ν_{i}^{2}) / E_{i}

where ν and E are Poisson’s ratio and elastic modulus of bone, respectively. The subscripts b and i represent bone and diamond indenter, respectively. β is indenter constant. Poisson’s ratio of bone ν_b= 0.3; Poisson’s ratio (ν_i) and elastic modulus (E_i) of the Berkovich diamond indenter are 0.07 and 1140 GPa, respectively. Indenter constant β = 1.034 [23].

Statistical analysis

One-way ANOVA was used to test the main effect of moderate treadmill exercise with different levels weight-bearing [36]. Data analysis was performed with SPSS 19.0 software, and the significance level was set to p < 0.05.

Results

Figure 3 shows the body weight of each rat group during the experiment. Increasing weight-bearing level showed no significant effects on body weight for all time points (p > 0.05). The cross-sectional area in the middle segment of femoral shaft, length of the left femur, and calf muscles weight of rats in each group are shown in Figs. 4, 5, 6, respectively. No significant differences in femur length and calf muscles weight were detected among groups (p > 0.05, Figs. 5, 6).

Fig. 3 — The body weight of each group during the experiment. Error bars represent standard deviation (SD). At T2, the 8-week exercise program started

Fig. 4 — The cross-sectional areas in the middle segment of femoral shaft at the end of experiment. Error bars represent SD. ^△Statistically different from the ex group (p < 0.05); *Statistically different from the eb5 group (p < 0.05); ^◇Statistically different from the eb12 group (p < 0.05); ^#Statistically different from the eb19 group (p < 0.05); ^□Statistically different from the eb26 group (p < 0.05)

Fig. 5 — The length of femur. Error bars represent SD

Fig. 6 — The calf muscles weight. Error bars represent SD