Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Apr 1.
Published in final edited form as: J Bone Miner Res. 2019 Jan 7;34(4):711–725. doi: 10.1002/jbmr.3648

Effects of High-Fat Diet and Body Mass on Bone Morphology and Mechanical Properties in 1100 Advanced Intercross Mice

Matthew J Silva 1,2,*, Jeremy D Eekhoff 2, Tarpit Patel 1,†, Jane P Kenney-Hunt 3,††, Michael D Brodt 1, Karen Steger-May 4, Erica L Scheller 5, James M Cheverud 3,6
PMCID: PMC6879418  NIHMSID: NIHMS1053070  PMID: 30615803

Abstract

Obesity is generally protective against osteoporosis and bone fracture. However, recent studies indicate that the influence of obesity on the skeleton is complex and can be detrimental. We evaluated the effects of a high-fat, obesogenic diet on the femur and radius of 1100 mice (males and females) from the Large-by-Small advanced intercross line (F34 generation). At 5 months age, bone morphology was assessed by microCT and mechanical properties by three-point bending. Mice raised on a high-fat diet had modestly greater cortical area, bending stiffness and strength. Size-independent material properties were unaffected by a high-fat diet, indicating that diet influenced bone quantity but not quality. Bone size and mechanical properties were strongly correlated with body mass. However, the increases in many bone traits per unit increase in body mass were less in high-fat diet mice than low-fat diet mice. Thus, although mice raised on a high-fat diet have, on average, bigger and stronger bones than low-fat fed mice, a high-fat diet diminished the positive relationship between body mass and bone size and whole-bone strength. The findings support the concept that there are diminishing benefits to skeletal health with increasing obesity.

More than one-third of the world’s population is overweight or obese (1). Individuals with higher body weight or body mass index (BMI) have increased bone mass or bone mineral density (BMD) (2–4), traits positively correlated with whole-bone strength (5–8). Furthermore, abundant data show that higher BMI is associated with lower fracture risk at the wrist, spine and hip (9–11). However, the influence of BMI on fracture risk is non-linear; increases in BMI from underweight to normal are strongly protective against fracture, while increases from overweight to obese are less protective (9). Thus, obesity may not be associated with weight-proportionate, positive effects on bone mass, size or strength (12).

Mice provide a model to study associations between body size, obesity and skeletal properties. Several studies have demonstrated positive associations between bone size, bone strength and body size (13–15). For example, a study of 662 NZBRF-F2 female mice revealed that bone size and strength were moderately correlated with body weight (r = 0.31–0.52) (15). However, the generalizability of these past findings is limited for several reasons: mice were selected for divergent bone size but not divergent body size, the influence of obesity was not addressed, and studies focused solely on the femur.

High fat feeding in rodents leads to obesity, and is consistently reported to have a detrimental effect on cancellous bone mass in mice (16–19). However, the effect of diet-induced obesity on cortical bone mass is less clear; studies have reported no change (16), increased (20,21) and decreased (17,19) cortical bone mass. The majority of these studies used male C57Bl/6 mice, a strain with relatively low bone mass and high susceptibility to diet-induced obesity; it is unclear how results may differ in female mice or in mice of different genetic backgrounds.

We sought to assess the influence of body weight and obesity on cortical bone in female and male mice having a range of body sizes due both to genetic variation and diet. Accordingly, we turned to the Large-by-Small advanced intercross (LG,SM AI) line, which was created from an initial cross of inbred strains chosen for large and small extremes of body size, respectively. In their F16 generation, there was substantial genetic variance and sexual dimorphism in obesity-related phenotypes and in the response to a high-fat diet (22). Moreover, a study using 97 LGXSM F16 recombinant inbred mice revealed considerable variation in bone size and strength, as well as significant associations between bone traits and fat mass (23). Thus, the LG,SM AI mouse population is an ideal resource to study variations in bone properties, body size and their interrelationship. We reared female and male mice from the F34 generation of LG,SM mice on low-fat or high-fat diets until skeletal maturity, and evaluated cortical bone morphology and mechanical properties of the femur and radius. We asked: 1) What is the effect of a high-fat, obesogenic diet on cortical bone traits in female and male LG,SM AI mice? 2) Do cortical bone traits depend on body mass, and are these relationships influenced by diet and/or sex?

METHODS

Mice

This study was conducted with approval of the Washington University Institutional Care and Use Committee (IACUC), and followed guidelines of the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals. Bones were obtained from 1,139 mice from 137 full-sib families of the F34 generation of the LG/J by SM/J AI line (Wustl:LG,SM-G34). The derivation of these animals was described in detail recently for a study unrelated to bone (24). In brief, the line originated from an intercross of LG/J female and SM/J male inbred mice (Jackson Laboratory, Bar Harbor). These founder strains were chosen based on their divergent body weight; LG/J mice represent a strain with high body weight and SM/J a strain with low body weight (25–27). F1 hybrid mice were intercrossed to produce F2 hybrids. The AI line was maintained by random, non-sibling matings of 50–75 mating pairs per generation from the F2 through the F34 generation.

At 3-weeks age, each family was evenly divided by sex and randomly assigned to diet. Mice were fed chow either relatively high in fat (42% calories from fat; Harlan Teklad catalog #TD88137) or low in fat (15% calories from fat; Research Diets catalog #D12284) as described previously (28) (Supplemental Table S1). They were housed up to 5 per cage on a 12:12 h light:dark cycle, with ad libitum access to food and water. Animals were euthanized at an average age of 24.7 weeks (SD = 2.1; range = 21.0 – 28.7), when they were skeletally mature (23,29). We determined 29 traits per mouse (Supplemental Table S2), as detailed below. Data were collected by blinded observers.

Whole-Body Measures

Mice were fasted for 4 hours and then anesthetized using sodium pentobarbital. Blood was collected via cardiac puncture and animals died by exsanguination. Serum was separated from whole blood by centrifugation and stored frozen. Serum leptin was determined by ELISA (Crystal Chem, Inc.). Immediately post mortem, we measured body mass, and dissected and weighed four fat pads – reproductive, inguinal, perirenal, and mesenteric (22). The sum of these was designated as the total fat pad mass, and taken as a measure of obesity. Unilateral femora and radii were dissected from each mouse, wrapped in saline-soaked gauze and stored at −20°C for later analysis.

Bone Morphology

Femoral and radial bone lengths were measured using calipers. Cross-sectional bone morphology was assessed at the mid-diaphysis of each femur and radius using microCT (μCT40, Scanco Medical) similar to a previous report (23) and consistent with published guidelines (30). Prior to scanning, bones were thawed at room temperature and embedded in 1.5% agarose. Twelve transverse slices were obtained spanning a 3 mm region of the mid-diaphysis (16 μm voxel size). Data were imported into ImageJ (NIH, Bethesda), and segmented into binary images using a threshold midway between the grayscale values of bone and background. Each set of femur images was rotated such that the medial-lateral bone axis was horizontal (x-axis), corresponding to the orientation of the neutral axis during subsequent bending testing. Images of the radii were similarly rotated to orient the neutral axis horizontally. We used a custom macro (Microsoft Excel) to compute standard bone morphological parameters (31): cortical bone area (Ct.Ar), total area (Tt.Ar), marrow area (Ma.Ar), polar moment of inertia (J), and average cortical thickness (Ct.Th). Values for each parameter were averaged over the 12 slices of each bone. After scanning, bones were stored at −20°C until mechanical testing.

Whole-Bone Mechanical and Tissue-Level Material Properties

Mechanical properties of each femur and radius were determined by three-point bending using standard techniques for assessing diaphyseal cortical bone (31). Tests were performed using a materials testing system (Instron Dynamight 8841). Bones were thawed in 0.9% phosphate buffered saline solution and kept moist during testing. They were placed on two support points (span = 7 mm), with the bone mid-point positioned beneath the upper point of the loading fixture. After applying a pre-load of −0.3 N, the loading point was displaced downward at 0.1 mm/sec until failure (fracture). Force and displacement data were collected at 250 Hz (LabView, National Instruments), and were visualized as force-displacement plots. From these we determined whole-bone mechanical properties: stiffness (K), ultimate force (Fult, also known as maximum load), post-yield displacement (PYD), and work-to-fracture (Wfx).

To estimate tissue-level (i.e., size-independent) material properties, we used simple beam theory equations and computed elastic modulus (E) and ultimate stress (σult). We report material properties for the radius only, because of the recognized limitations of this approach for estimating femoral material properties (31,32).

Statistical Analysis

Analyses were conducted using commercial software (SAS Studio 3.4) with a significance level of p < 0.01. Mice were categorized into four groups based on sex and diet: female low-fat diet (FL), female high-fat diet (FH), male low-fat diet (ML), or male high-fat diet (MH). A set of eight histograms for each of the 29 traits, sub-divided by sex, by diet, and by sex and diet were examined to assess normality. For nine traits (including all seven of the whole-body traits), the distribution appeared non-normal, and we applied a natural log, inverse or square root transformation to normalize the data. [A prefix of (L), (i) or (S) denotes transformed data.] Descriptive statistics are presented as non-transformed values, in either data tables (mean ± SD) or box and whisker plots (horizontal bars depict the minimum, 25th percentile, median, 75th percentile, and maximum values). Where percent differences are stated, they are computed based on non-transformed, mean values. Statistical analyses described below were conducted on data after any necessary transformations.

Bivariate scatter plots for each independent combination of variables were examined to identify extreme values and confirm the appropriateness of using linear models to assess relationships between traits. Extreme values (judged to be either physically or biologically implausible) of one or more traits were identified for 26 mice. These mice there were removed from the dataset, leaving 1113 mice for analysis (FL: n = 274; FH: n = 282; ML: n = 274; MH: n = 283). There were some incomplete data due to technical issues or user error, resulting in a total sample of N = 1087–1113 for the different traits. Tables of Pearson correlation coefficients were produced for the dataset as a whole (Supplemental Table S3) and for each group to determine the strength of the pairwise, linear relationships between traits. First, it was noted that no traits were dependent on age at euthanasia (|r| ≤ 0.17), and consequently age was judged not to be a significant confounder for the data and was not considered further. Next, the correlation tables were examined to determine which traits were most representative of body size. Body mass was highly correlated with the individual and total fat pad masses (r ≥ 0.81), and therefore body mass was used as the primary trait describing body size. Also, because each individual fat pad mass was highly correlated with the total fat pad mass (r ≥ 0.90), only the total fat pad mass was used to represent adiposity in subsequent analyses.

To address our first research question, a two-way ANOVA was performed to determine the effects of sex and diet and their possible interaction on each trait. Non-significant (ns) effects were removed from the model (although main effects were retained regardless of significance if the interaction was significant). Levene’s test was used to test for homogeneity of variance between the groups. Where differences between variances were significant, a one-way Welch’s ANOVA was performed and compared to the results of the standard ANOVA to confirm that the heterogeneity of variance did not significantly alter the results. The Tukey-Kramer post-hoc test was used to compare differences in least squares means between groups.

To address our second research question, analysis of covariance (ANCOVA) was performed for each bone trait using body mass as a continuous covariate and diet and sex as independent, categorical variables. (Details of the linear ANCOVA model are presented in the Supplemental Materials section.) Initially, each independent variable and all possible interactions were included in the model. Then, various selection techniques including backwards, stepwise, and manual selection were utilized to find a number of potential models to fit the data. For a model to be considered for further analysis, each term in the model must have been significant, with the exception necessary to preserve model hierarchy, i.e., if a significant interaction term was in the model, the main effects for that interaction remained regardless of significance. Of all models that met this requirement, the model with the highest adjusted R2 value was chosen as the final model that was most descriptive of the data. ANCOVA results are presented as scatter plots of trait value vs. body mass, with regression lines representing the final model equations. Lastly, predicted values were computed based on the final model equations evaluated at the overall mean body mass (i.e., mean of all 1113 mice). Differences between the predicted values for each study group at the overall mean body mass were assessed using the Tukey-Kramer post-hoc test that adjusts for multiple comparisons.

Results

High-Fat Diet Leads to Obesity in Female and Male LG,SM AI Mice

Mice fed a high-fat diet had significantly higher body mass, total fat pad mass, and serum leptin than mice fed a low-fat diet (Fig. 1, Supplemental Table S4), consistent with results from an earlier generation (F16) of LG,SM AI mice (22). The diet-related increases in body mass (30–35% greater mean values for high-fat vs. low-fat diet) and leptin (100–120% greater) were similar in females and males. By contrast, total fat pad mass was increased by high-fat diet more in females (160%) than in males (100%). This sex-diet interaction (p = 0.007) was mostly because females on a low-fat diet were leaner than males (8.9% vs 12.0% total fat pad mass/body mass, p < 0.001), a difference that was nullified in high-fat diet mice (17.5 vs. 18.4%, p = 0.08).

Figure 1:

Figure 1:

Open in a new tab

Mice raised on a high-fat diet had significantly greater (a) body mass, (b) total fat pad mass, and (c) serum leptin compared to mice raised on a low-fat diet. Males had greater values of these traits than females. Data are from 1100 LG,SM AI mice.

We used simple correlation and analysis of covariance (ANCOVA) to examine relationships between whole-body traits. Total fat pad mass was highly correlated with body mass (r = 0.90, Suppl. Table S3), and serum leptin was highly correlated with both body mass (r = 0.70) and total fat pad mass (r = 0.74). ANCOVA revealed that the linear relationship between total fat pad mass and body mass was independent of diet (Suppl. Fig. S1). In other words, obesity scaled with body size whether increased body size was due to genetic variation or to a high-fat diet. On the other hand, the fat mass-body mass relationship was sex-dependent. Females had a greater increase in total fat pad mass per unit increase in body mass than males, consistent with the greater relative accrual of fat by females on a high-fat diet.

To analyze the influence of body size on bone traits (presented below), we selected body mass as a covariate. Of the whole-body traits, body mass had the greatest number of moderate-to-strong correlations with bone traits (|r| > 0.3 for 15 of 22 traits). By contrast, total fat pad mass was moderately-to-strongly correlated with 9 of 22 bone traits, and serum leptin with only 3 of 22 bone traits (Suppl. Table S3). In summary, mice raised on a high-fat diet have increased body mass and obesity compared to low-fat fed mice, and body mass is the whole-body trait that correlates most strongly with bone traits.

Female and Male Mice Fed a High-Fat Diet Have Modestly Increased Cortical Area in the Femur and Radius

We used microCT to assess bone morphological traits, and found that mice fed a high-fat diet had increased cortical bone size, although the magnitude of the diet effect was much less than it was for body size. Diet affected femora and radii differently. In the femur, mice fed a high-fat diet had modestly greater cortical area than mice fed a low-fat diet (3–4%; Fig. 2a). This effect was not due to periosteal expansion, but was related to endocortical infilling; total area was unaffected by diet (Fig. 2b), whereas marrow area was less in mice fed a high-fat diet (-5%; Fig. 2c). Consistent with increased cortical area, cortical thickness was greater in high-fat diet mice (6–7%; Fig. 2d). In addition, consistent with the lack of a diet effect on total area, moment of inertia was unaffected by diet (Fig. 2e). Lastly, femora of mice fed a high-fat diet were slightly longer (1%; Suppl. Fig. S2a) than mice fed a low-fat diet. Male mice had larger femora by all measures than female mice, but the effect of diet was similar in males and females, i.e., there was no significant sex-diet interactions for femoral morphology.

Figure 2:

Figure 2:

Open in a new tab

Femurs of mice raised on a high-fat diet had slightly larger (a) cortical area and (d) cortical thickness, and smaller (c) marrow area than mice raised on a low-fat diet. Diet did not influence (b) total area, or (e) moment of inertia. Males had larger values of all femur morphology traits than females. Results are from 1100 LG,SM AI mice, analyzed by ANOVA, without accounting for body mass.

Like the femur, the radius from mice fed a high-fat diet had modestly greater cortical area than low-fat fed mice (1–3%; Fig. 3a). In contrast to the femur, this effect occurred via slight periosteal expansion as evidenced by increased total area (1–2%) in the radius of high-fat diet mice (Fig. 3b), with no effect on marrow area (Fig. 3c). Consistent with these effects, cortical thickness (1–3%; Fig. 3d) and moment of inertia (3–5%; Fig. 3e) were modestly greater in the high-fat diet groups. Lastly, radii of mice fed a high-fat diet were slightly longer (1%; Suppl. Fig. S2b) than mice fed a low-fat diet, similar to the femur. Also like the femur, the radius of male mice was larger than of female mice, but diet effects were similar in female and male mice, with no significant sex-diet interaction for radial morphology. In summary, a high-fat diet modestly increases cortical area, in association with smaller marrow area in the femur and greater total area in the radius.

Figure 3:

Figure 3:

Open in a new tab

Radii of mice raised on a high-fat diet had slightly larger (a) cortical area, (b) total area, (d) cortical thickness, and (e) moment of inertia, than mice raised on a low-fat diet. Diet did not influence (c) marrow area. Males had larger values of all radial morphology traits than females. Results are from 1100 LG,SM AI mice, analyzed by ANOVA, without accounting for body mass.

Female and Male Mice Fed a High-Fat Diet Have Modestly Increased Whole-Bone Mechanical Properties but No Changes in Material Properties

We next evaluated mechanical properties (traits) using bending tests. Whole-bone mechanical properties (Suppl. Table S2) are measured directly from the three-point bending tests and reflect the behavior of the bone as a structure. Mice fed a high-fat diet had higher femoral bending stiffness (5–7%; Fig. 4a), ultimate force (6–8%; Fig. 4b), and work-to-fracture (8–15%; Fig. 4d) than mice fed a low-fat diet. Like the femur, the radius from high-fat fed mice had higher stiffness (3–6%; Fig. 5a), although radial ultimate force (Fig. 5b) and work-to-fracture (Fig. 5d) were not significantly affected by diet.

Figure 4:

Figure 4:

Open in a new tab

Femurs of mice raised on a high-fat diet had modestly greater (a) stiffness, (b) ultimate force, and (d) work-to-fracture. Diet did not influence (c) post-yield displacement. Bones from males had greater values of stiffness and ultimate force, but lesser post-yield displacement and equivalent work-to-fracture compared to females. Results are from 1100 LG,SM AI mice, analyzed by ANOVA, without accounting for body mass.

Figure 5:

Figure 5:

Open in a new tab

Radii of mice raised on a high-fat diet had modestly increased (a) stiffness; no other radial whole-bone mechanical or bone material properties was affected by diet (b-f). Bones from males had greater stiffness, ultimate force, work-to-fracture, modulus and ultimate stress, but post-yield displacement was not different compared to females. Diet effects were similar in female and male mice. Results are from 1100 LG,SM AI mice, analyzed by ANOVA, without accounting for body mass.

We also estimated the bone material properties for the radius (Suppl. Table S2). These properties are adjusted for bone size and represent intrinsic properties of the cortical bone. Based on ANOVA of estimated elastic modulus (E) and ultimate stress (σult), there was no effect of dieton radial material properties (Figs. 5e–f). In addition, diet did not influence post-yield displacement of either the femur (Fig. 4c) or radius (Fig. 5c). This property, although obtained directly from the whole-bone test, is generally independent of bone size and reflects material behavior (31,33).

Compared to female bones, male bones (femur and radius) had higher values of stiffness and ultimate force, consistent with their bigger size. Nevertheless, the effects of diet were similar in female and male mice, with no significant sex-diet interactions for any femoral or radial mechanical properties. In summary, a high-fat diet enhanced whole-bone mechanical properties of the femur, and to a lesser degree, the radius. High-fat diet did not influence post-yield displacement or estimated material properties.

The Positive Relationship between Cortical Morphology and Body Mass is Diminished by a High-Fat Diet

Because bone traits are often associated with body size, we next examined relationships between bone traits and body mass, and whether these relationships depended on diet and/or sex. Femur morphology and body mass were positively correlated, i.e., bigger mice generally had bigger femurs. Simple, bivariate correlations were strong between body mass and cortical area, total area, and moment of inertia (r = 0.61–66; Suppl. Table S3), and were moderate versus marrow area, cortical thickness, and femoral length (r = 0.41–0.49). ANCOVA models that accounted for the effects of body mass and diet (and sex) revealed that diet (either directly or through an interaction) significantly influenced all six femur morphology traits (Fig. 6; Suppl. Fig. S2c). Importantly, there was a significant, negative interaction between body mass and diet for five of six traits, which means that bone traits of mice on a high-fat diet were less positively influenced by increases in body size than mice on a low-fat diet. For example, considering total area as a trait representative of femoral size, mice fed a high-fat diet had smaller increases in total area per unit increase in body mass compared to mice fed a low-fat diet (diet*BM, p < 0.0001; Fig. 6b). Moreover, based on model predictions evaluated at the overall mean body mass, the total area of femurs from high-fat diet mice was 5–10% less than what would be expected for femurs from low-fat diet mice of comparable size (Table 1). Similar relative effects of diet were noted for cortical area, marrow area, moment of inertia and length (Figs. 6a,c,e; Suppl. Fig. S2c). The ANCOVA models predicted that the negative influence of high-fat diet on body-size adjusted properties was stronger in females than males (Table 1). The only trait that did not follow this pattern was cortical thickness, which was slightly (2%) greater for mice fed a high-fat diet, and did not have a significant diet*BM interaction (Fig. 6d). In summary, femoral size was positively correlated with body mass, but a high-fat diet modestly diminished this positive effect for most measures of bone size.

Figure 6:

Figure 6:

Open in a new tab

Femur morphology traits vs. body mass (log transformed) for 1100 LG,SM AI mice. ANCOVA models that account for body mass, diet and sex and their interactions explain 24–51% of the trait variance. Note that the slope of the regression lines for low-fat diet (LFD) groups are greater than for high-fat diet (HFD) groups for four of five traits (a-c, e; diet*BM, p < 0.001), whereas the slopes for the female and male subgroups are not different (sex*BM, ns). Sex*diet*BM interaction is not significant. For Figures 6–9, regression lines depict results of final models; four lines are shown when diet and sex (or their interaction) were significant; two lines are shown when diet or sex alone was significant; one line is shown when neither diet nor sex are significant, and BM is the only term in the model. (FL: Female, LFD; FH: Female, HFD; ML: Male, LFD; MH: Male, HFD)

Table 1.

Predicted values for bone morphology traits by sex and by diet, based on analysis of covariance (ANCOVA) models evaluated at the overall mean body mass.

Bone Trait Female Male
Low-Fat (FL) High-Fat (FH) Difference High-Fat vs. Low-Fat Low-Fat (ML) High-Fat (MH) Difference High-Fat vs. Low-Fat
Femur Ct.Ar (mm2) 0.98 0.91 −7% 0.97 0.95 ns
Tt.Ar (mm2) 1.73 1.55 −10% 1.74 1.65 −5%
Ma.Ar (mm2) 0.75 0.63 −15% 0.77 0.70 −9%
Ct.Th* (mm) 0.245 0.250 +2% 0.245 0.250 +2%
J (mm4) 0.40 0.33 −18% 0.41 0.38 −7%
Length (mm) 16.3 16.1 −2% 16.1 16.0 ns
Radius Ct.Ar (mm2) 0.30 0.28 −5% 0.29 0.29 ns
Tt.Ar (mm2) 0.35 0.33 −5% 0.35 0.34 ns
Ma.Ar* (mm2) 0.055 0.049 −11% 0.055 0.049 −11%
Ct.Th* (mm) 0.21 0.21 ns 0.21 0.21 ns
J (mm4) 0.020 0.018 −10% 0.020 0.019 ns
Length (mm) 11.9 11.7 −2% 11.9 11.9 ns
Open in a new tab

ns: not significant, p > 0.01, Tukey-Kramer post hoc test

*

Female vs. male not different; data pooled

Similar to the femur, radius morphology and body mass were positively related, although the correlations were less strong. Simple, bivariate correlations between body mass and cortical area, total area, moment of inertia, and length were moderately strong (r = 0.46–0.50), but were weak versus marrow area and cortical thickness (r = 0.15–0.36). Similar to the femur, ANCOVA analysis revealed that diet (either directly or through an interaction) had a significant influence for five of six radial traits (Fig. 7; Suppl. Fig. S2d). Importantly, there was a significant, negative interaction between body mass and diet for three traits. For example, mice fed a high-fat diet had smaller increases in radius total area per unit increase in body mass compared to mice fed a low-fat diet (diet*BM, p < 0.0001; Fig. 7b). These effects were also noted for cortical area and radius length (Fig. 7a; Suppl. Fig. S2d). Again similar to the femur, the ANCOVA models predicted a greater negative influence of high-fat diet for females than males (Table 1). An exception was cortical thickness, which was independent of diet (Fig. 7d). In summary, bigger mice generally had bigger bones, and thus a high-fat diet on average was associated with increased bone size (Figs. 3, 4). However, mice fed a high-fat diet showed smaller increases in bone size per increase in body size compared to low-fat fed mice (Figs. 6, 7), with effects more strongly observed in the femur than the radius, and more so in females than males.

Figure 7:

Figure 7:

Open in a new tab

Radius morphology traits vs. body mass (log transformed) for 1100 LG,SM AI mice. ANCOVA models that account for body mass, diet and sex and their interactions explain 2–27% of the variance in these traits. Note that the slope of the regression lines for low-fat diet (LFD) groups are greater than for high-fat diet (HFD) groups for two of five traits (a-b; diet*BM, p < 0.001), whereas the slopes for the female and male subgroups are not different (sex*BM, ns). Sex*diet*BM interaction is not significant. (FL: Female, LFD; FH: Female, HFD; ML: Male, LFD; MH: Male, HFD)

Whole-Bone Mechanical Properties Correlate Positively with Body Size but Material Properties Do Not

Finally, we examined the influence of body size on bone mechanical properties, and found complex relationships between mechanical properties, body size, and diet. Most whole-bone mechanical properties of the femur were positively correlated with body size (Fig. 8). Bivariate correlations versus body mass were moderate-to-strong for femoral stiffness and ultimate force (r = 0.36–0.63), but weak for work-to-fracture (r = 0.17). Post-yield displacement was independent of body size. ANCOVA models that accounted for body size revealed a significant diet effect (diet, p < 0.01) for three of six femoral mechanical properties, although in contrast to the femur morphology results, no models contained significant diet-body mass interactions (diet*BM, ns; Fig. 8). Based on predicted values from the ANCOVA model, femurs from high-fat diet mice had a modestly lower ultimate force than femurs from similarly sized low-fat diet mice (-4%, Table 2, Fig. 8b), which is consistent with the ANCOVA model predictions for femur size (Table 1).

Figure 8:

Figure 8:

Open in a new tab

Femur mechanical properties vs. body mass (log transformed) for 1100 LG,SM AI mice. ANCOVA models that account for body mass, diet and sex and their interactions explain 2–42% of the trait variance. Note that diet affected only ultimate force (b). Moreover, while there is a significant body mass effect for three of four traits, this effect did not depend on diet (diet*BM, ns). Sex*diet*BM interaction is not significant. (FL: Female, LFD; FH: Female, HFD; ML: Male, LFD; MH: Male, HFD)

Table 2.

Predicted values for bone mechanical properties by sex and by diet, based on analysis of covariance (ANCOVA) models evaluated at the overall mean body mass.

Bone Trait Female Male
Low-Fat (FL) High-Fat (FH) Difference High-Fat vs. Low-Fat Low-Fat (ML) High-Fat (MH) Difference High-Fat vs. Low-Fat
Femur K*† (N/mm) 161 161 ns 161 161 ns
F_ult (N) 27.6 26.4 −4% 27.8 26.6 −4%
(i)PYD† (1/mm) 0.22 0.22 ns 0.19 0.19 ns
(S)W_fx† (Nmm)1/2 2.95 2.95 ns 2.83 2.83 ns
Radius K (N/mm) 25.7 23.1 −10% 25.1 25.2 ns
F_ult (N) 5.5 5.0 −9% 5.4 5.2 ns
PYD*† (mm) 0.64 0.64 ns 0.64 0.64 ns
W_fx† (Nmm) 2.9 2.9 ns 3.1 3.1 ns
E† (GPa) 22.9 22.9 ns 24.0 24.0 ns
σ_ult† (MPa) 377 377 ns 387 387 ns
Open in a new tab

ns: not significant, p > 0.01, Tukey-Kramer post hoc test

*

Female and male not different; data pooled

†

Low-fat and High-fat not different; data pooled

Similar to the femur, there was a positive correlation between body size and whole-bone mechanical properties of the radius, with the exception of post-yield displacement (Fig. 9). Bivariate correlations versus body mass were moderate for radial stiffness and ultimate force (r = 0.41–0.42), but weak for work-to-fracture (r = 0.22). ANCOVA models that accounted for body size revealed that diet significantly affected ultimate force directly (diet, p = 0.0087; Fig. 9b) and through its interaction with body mass (diet*BM, p =0.0033); similar trends were observed for stiffness (0.01 < p < 0.02 Fig. 9a). Based on ANCOVA models evaluated at the overall mean body mass, radii from high-fat diet female mice had 9–10% lower stiffness and ultimate force than from low-fat diet female mice (Table 2), although the radii from male mice showed no diet effect. Notably, diet did not significantly affect any other whole-bone mechanical properties or material properties (Wfx, PYD, E, σult; Figs. 9c–f). In summary, bones from bigger mice were generally stiffer and stronger. Some whole-bone mechanical properties (e.g., ultimate force) from mice fed a high-fat diet were less than from similarly sized mice fed a low-fat diet, although post-yield displacement and bone material properties were independent of body size and unaffected by diet.

Figure 9:

Figure 9:

Open in a new tab

Radius mechanical properties vs. body mass (log transformed) for 1100 LG,SM AI mice. ANCOVA models that account for body mass, diet and sex and their interactions explain 0–21% of the trait variance. Note that diet affected stiffness (a) and ultimate force (b), and that the slopes of the regression lines for these traits are greater for the low-fat diet (LFD) groups than for high-fat diet (HFD) groups (a-b; diet*BM, p < 0.001). Neither diet nor body mass influenced post-yield displacement (c), or radius material properties (e-f). (FL: Female, LFD; FH: Female, HFD; ML: Male, LFD; MH: Male, HFD)

Discussion

We examined cortical bone traits of the femur and radius from a large set of skeletally mature female and male LG,SM AI mice raised on low- or high-fat diets. A high-fat diet significantly increased body mass and total fat pad mass (a measure of obesity) in both sexes, with females showing a greater relative increase in fat mass than males. We first examined the effects of diet on bone traits using ANOVA without accounting for the influence of body mass. High-fat diet had a positive effect on most measures of femur size and whole-bone mechanical properties, and also positively affected radius size and stiffness. Thus, based on analyses that did not account for body size, the overall effect of a high-fat diet on bone size and whole-bone mechanical properties was positive, or, at worst, neutral. Next, we considered body mass effects. All bone morphology traits and most whole-bone mechanical properties were positively correlated with body mass, i.e., bigger mice on average had bigger, stronger bones. Using ANCOVA to examine the effects of diet and sex and body mass revealed a complex picture. While bone size and whole-bone mechanical properties increased with body size, the increase in most bone traits per unit increase in body mass was less in high-fat diet mice than low-fat diet mice. Thus, although mice raised on a high-fat diet have, on average, bigger and stronger bones than low-fat fed mice, a high-fat diet diminished the positive relationship between body mass and bone size and whole-bone strength.

In contrast to the significant effects of high-fat diet and body mass on bone size and whole-bone mechanical properties (e.g., stiffness, ultimate force), these factors did not influence estimated bone material properties (modulus, ultimate stress). The standard approach we used to estimate material properties from bending tests and beam theory has recognized limitations when applied to rodent long bones, especially mouse femora (31,32,34,35). We chose to not report estimated material properties for the femur, but do report properties for the radius, which should be more accurate (32). The values of elastic modulus (overall mean = 23.5 GPa; Fig. 5e) are in the range expected for mouse cortical bone, although we note that there remains a size bias whereby larger bones tend to have lower modulus values (Suppl. Fig. S3a–b). Another issue is that beam theory does not fully account for the effects of yielding when estimating ultimate stress. Analysis of the relationship between ultimate force and section modulus suggests that post-yield behaviors are similar between groups, although radii from the male high-fat diet group may have modestly enhanced behavior compared to the other groups (Suppl. Fig. S3). Finally, neither diet nor body mass significantly affected post-yield displacement of the radius or the femur; PYD is measured directly from the whole-bone test but is generally independent of bone size and reflects the relative brittleness of bone (31,33). Thus, our overall findings indicate that a high-fat diet and body mass affect bone at the structural (whole-bone) level, but not at the material (bone quality) level.

Our findings have relevance to clinical data linking BMI and fracture risk. We found a positive correlation between body mass and bone size in mice, consistent with human data that body weight or BMI are correlated positively with BMD (2–4). Moreover, we observed positive correlations between body mass and whole-bone mechanical properties related to fracture resistance (ultimate force, work-to-fracture), a finding consistent with the negative correlations between BMI and fracture risk in humans (9–11). Despite these positive correlations between bone traits and body size, our findings also support the concept of diminishing returns with increasing obesity (9,12). In particular, a modest (~5–10%) negative effect of high-fat diet on bone size and strength is evident after accounting for body size (Tables 1, 2). Moreover, with increasing body size, mice raised on a high-fat diet have a smaller increase in bone size than mice raised on a low-fat diet. To illustrate this phenomenon, we plotted data from four individual female mice whose values of femur total area and body mass fall near their respective ANCOVA model regression lines (Fig. 10). A small, high-fat diet fed mouse had an 8% smaller bone size than an equally-sized mouse raised on a low-fat diet, and the difference was 13% when comparing larger mice. These data indicate that, for acquiring an optimal set of skeletal traits, it is preferable to achieve a large body size from genetic influences rather than from a high-fat diet. Nonetheless, despite this relative “inefficiency” in bone size accrual linked to a high-fat diet, a larger body size, even achieved on a high-fat diet, appears beneficial for whole-bone (diaphyseal) size and strength.

Figure 10:

Figure 10:

Open in a new tab

Effect of diet on the relationship between femoral Tt.Ar (a representative measure of bone size) and body mass in LG,SM AI mice. The four data points shown are from individual females whose values fall close to their respective regression lines: a smaller mouse (BM = 21 g) raised on a low-fat diet, and an equally small mouse raised on a high-fat diet; a larger mouse (BM = 40.4 g) raised on a low-fat diet, and an equally large mouse raised on a high-fat diet. The regression lines shown are for the entire set of female mice (from Fig. 6b). For equivalent body mass, the average mouse raised on a high-fat diet (HFD) has a smaller bone size than its counterpart raised on a low-fat diet (LFD), and this difference increases with body mass. Similar findings were observed in male mice, as well as in the radius of both sexes.

The documented effects of high-fat diet on cortical bone in rodents are complex, with positive, negative, and neutral results reported (reviewed in (36)). Three key variables dictate the impact of high-fat diet feeding on bone: age/duration of intervention, composition of diet, and secondary development of metabolic complications (i.e., hyperglycemia, insulin resistance). In our study, we examined dietary effects during growth – from weaning to adulthood (18- to 26-weeks on diet). Within this timeframe, both positive (37) and negative (19) effects of high-fat diet on cortical bone have been reported in male C57BL/6J mice. The main difference between these two prior studies was the use of 45% high-fat diet with diacylglycerol enrichment in the former (37) versus 60% lard/soybean oil in the latter (19). The dietary composition we used (42% milk fat) was comparable to the first study and, like that report, we demonstrated positive absolute effects of high-fat diet on cortical bone.

Overall, our findings fit the mechanistic paradigm described by Lecka-Czernik et al. where the effects of high-fat diet are bi-phasic (21).There is an initial beneficial effect of obesity on bone, likely due to the anabolic effects of increased mechanical loading (21) and perhaps also to elevated IGF-1 (38). The former is particulary relevant to our study because the anabolic response to skeletal loading is greatest during growth (2,39). However, a secondary phase with decreased bone formation then occurs due to development of metabolic complications including systemic inflammation. Although we did not assess metabolic outcomes here, in mice from the F16 generation of the LG,SM AI line, a high-fat diet led to reduced glucose tolerance and elevated serum insulin (22,40). Others have reported that trabecular bone loss induced by a high-fat diet is modulated via the effects of inflammatory cytokines TNF-α and IGF-1 (41,42). These competing effects – increased bone formation due to increased body weight versus a pre-diabetic condition that may inhibit bone formation and/or enhance resorption – are consistent with our overall finding that a high-fat diet was good for bone in absolute terms (i.e., associated with absolute increases in bone size and whole-bone strength), but was relatively detrimental if we account for body size. We note that these findings may be different in older mice that become obese as adults, or that continue to consume a high-fat diet after reaching skeletal maturity.

Leptin has been widely studied for its skeletal effects (43). We observed that increased circulating leptin correlated strongly with increased body mass and fat mass (r = 0.67–0.74), consistent with previous results (44,45). On the other hand, serum leptin correlated weakly with most bone traits (Suppl. Table S3). Only three traits correlated moderately (r > 0.3) with leptin: femur cortical area (r = 0.33), moment of inertia (r = 0.31) and ultimate force (r = 0.35). By comparison, these traits correlated more strongly with body weight (r = 0.63–0.66). Thus, our results indicate that in a murine population where body mass, fat mass and serum leptin varied across a wide physiologic range (leptin range: 0.04–53.6 ng/ml), leptin was only weakly associated with skeletal traits, and body weight was a better correlate with measures of bone size and strength.

Our study is notable in examining the associations between body size and bone morphological and mechanical traits in a large sample of mice having a wide variation in body size. Two previous studies assessed similar phenotypic traits in large sets of hybrid mice (14,15), although the initial progenitor strains for those studies had relatively small differences in body weight (13% for C57Bl/6 vs. DBA/2 (14), and 10% for NZB vs. RF (15)). By contrast, LG/J and SM/J mice differ in body weight by 112% (data from 13-week old males (46)), and the LG,SM intercross (along with the two diet conditions) produced a wide variation in body size (BM range: 14.4 – 82.3 g), and bone traits (e.g., femur Ct.Ar: 0.61 – 1.49 mm2; femur ultimate force: 12.4 – 52.8 N). These conditions provide a powerful experimental model to examine the influence of body size on bone traits. Future analyses of this population will allow identification of quantitative trait loci (QTLs) related to bone size, mechanical properties, and their interrelationships with body size and obesity.

Another noteworthy aspect of our study is the assessment of body mass and diet effects on both the femur and radius. While the femur is substantially larger and has a 5-fold greater ultimate force than the radius, they are both weight-bearing long bones. Caged mice spend most of their active time in activities involving all four limbs (47), and during locomotion the forelimb and hindlimb experience comparable peak ground reaction forces (48,49). Thus, from a functional adaptation perspective, it is not surprising that morphological and mechanical properties of both bones are positively correlated with body mass, and that many of the relative effects of diet (and sex) are comparable in the femur and radius. A notable difference between the two bones is their response to high-fat diet. In the femur, cortical area was greater in high-fat diet due to a smaller marrow area, whereas in the radius it was greater due to larger total (periosteal) area (Supplemental Fig. S4). The significance of this envelope-specific difference remains to be determined.

Conclusions

In summary, we assessed the effects of diet and body mass on the morphological and mechanical traits of long bones of 1100 adult F34 LG,SM AI mice. Mice had a wide variation in body mass (5.7 fold from the smallest to the largest), bone size (2.4 fold) and bone strength (4.3 fold). Female and male mice raised on a high-fat diet had modestly greater (approx. +5%) cortical bone area and thickness, which was associated with a smaller marrow area in the femur, and a larger periosteal area in the radius. High-fat diet also enhanced measures of whole-bone stiffness and strength in the femur and, to a lesser extent, the radius. Size-independent material properties were unaffected by a high-fat diet, indicating that diet influenced cortical bone quantity but not bone quality. Bone size and mechanical properties were strongly correlated with body mass, i.e., bigger mice on average had bigger and stronger bones. Using ANCOVA to examine the combined effects of body size, diet and sex revealed a negative relative effect of high-fat diet, i.e., the increases in many bone traits per unit increase in body mass were less in high-fat diet mice than low-fat diet mice. Nonetheless, despite the negative influence that high-fat diet had on the bone size versus body size relationship, the absolute effects of a high-fat diet on bone traits was mostly positive, or at worst, neutral. We conclude that mice raised on a high-fat diet have, on average, bigger and stronger bones, but they achieve these traits via a slightly less efficient accrual of bone mass compared to mice raised on a low-fat diet.

Supplementary Material

Table S3
supp figs
supp tabs
suppl methods

Acknowledgements

Funded by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (R01DK075112) and the National Institute of Arthritis Musculoskeletal and Skin Diseases (R01AR047867, P30AR057235). We thank Dr. E. Ann Carson and Ms. Bing Wang for their contributions to data collection and analysis.

Footnotes

Disclosures

The authors state that they have not disclosures.

Supplemental Materials

This submission included 4 Supplemental Tables and 3 Supplemental Figures

References

  • 1.Hruby A, Hu FB. The Epidemiology of Obesity: A Big Picture. Pharmacoeconomics. 2015;33(7):673–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Moro M, van der Meulen MC, Kiratli BJ, Marcus R, Bachrach LK, Carter DR. Body mass is the primary determinant of midfemoral bone acquisition during adolescent growth. Bone. 1996;19(5):519–26. [DOI] [PubMed] [Google Scholar]
  • 3.Ishii S, Cauley JA, Greendale GA, et al. Pleiotropic effects of obesity on fracture risk: the Study of Women’s Health Across the Nation. J Bone Miner Res. 2014;29(12):2561–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Evans AL, Paggiosi MA, Eastell R, Walsh JS. Bone density, microstructure and strength in obese and normal weight men and women in younger and older adulthood. J Bone Miner Res. 2015;30(5):920–8. [DOI] [PubMed] [Google Scholar]
  • 5.Lotz JC, Hayes WC. The Use of Quantitative Computed-Tomography to Estimate Risk of Fracture of the Hip from Falls. Journal of Bone and Joint Surgery-American Volume. 1990;72a(5):689–700. [PubMed] [Google Scholar]
  • 6.Moro M, Hecker AT, Bouxsein ML, Myers ER. Failure load of thoracic vertebrae correlates with lumbar bone mineral density measured by DXA. Calcif Tissue Int. 1995;56(3):206–9. [DOI] [PubMed] [Google Scholar]
  • 7.Bouxsein ML, Courtney AC, Hayes WC. Ultrasound and densitometry of the calcaneous correlate with the failure loads of cadaveric femurs. Calcif Tiss Int. 1995;56(2):99–103. [DOI] [PubMed] [Google Scholar]
  • 8.Ebbesen EN, Thomsen JS, Beck-Nielsen H, Nepper-Rasmussen HJ, Mosekilde L. Age- and gender-related differences in vertebral bone mass, density, and strength. J Bone Miner Res. 1999;14(8):1394–403. [DOI] [PubMed] [Google Scholar]
  • 9.De Laet C, Kanis JA, Oden A, et al. Body mass index as a predictor of fracture risk: a meta-analysis. Osteoporos Int. 2005;16(11):1330–8. [DOI] [PubMed] [Google Scholar]
  • 10.Premaor MO, Compston JE, Fina Aviles F, et al. The association between fracture site and obesity in men: a population-based cohort study. J Bone Miner Res. 2013;28(8):1771–7. [DOI] [PubMed] [Google Scholar]
  • 11.Compston JE, Flahive J, Hosmer DW, et al. Relationship of weight, height, and body mass index with fracture risk at different sites in postmenopausal women: the Global Longitudinal study of Osteoporosis in Women (GLOW). J Bone Miner Res. 2014;29(2):487–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Palermo A, Tuccinardi D, Defeudis G, et al. BMI and BMD: The Potential Interplay between Obesity and Bone Fragility. Int J Environ Res Public Health. 2016;13(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Brochmann EJ, Duarte ME, Zaidi HA, Murray SS. Effects of dietary restriction on total body, femoral, and vertebral bone in SENCAR, C57BL/6, and DBA/2 mice. Metabolism. 2003;52(10):1265–73. [DOI] [PubMed] [Google Scholar]
  • 14.Lang DH, Sharkey NA, Lionikas A, et al. Adjusting data to body size: a comparison of methods as applied to quantitative trait loci analysis of musculoskeletal phenotypes. J Bone Miner Res. 2005;20(5):748–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Wergedal JE, Ackert-Bicknell CL, Tsaih SW, et al. Femur mechanical properties in the F2 progeny of an NZB/B1NJ x RF/J cross are regulated predominantly by genetic loci that regulate bone geometry. J Bone Miner Res. 2006;21(8):1256–66. [DOI] [PubMed] [Google Scholar]
  • 16.Cao JJ, Gregoire BR, Gao H. High-fat diet decreases cancellous bone mass but has no effect on cortical bone mass in the tibia in mice. Bone. 2009;44(6):1097–104. [DOI] [PubMed] [Google Scholar]
  • 17.Fujita Y, Watanabe K, Maki K. Serum leptin levels negatively correlate with trabecular bone mineral density in high-fat diet-induced obesity mice. J Musculoskelet Neuronal Interact. 2012;12(2):84–94. [PubMed] [Google Scholar]
  • 18.Inzana JA, Kung M, Shu L, et al. Immature mice are more susceptible to the detrimental effects of high fat diet on cancellous bone in the distal femur. Bone. 2013;57(1):174–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Scheller EL, Khoury B, Moller KL, et al. Changes in Skeletal Integrity and Marrow Adiposity during High-Fat Diet and after Weight Loss. Frontiers in endocrinology. 2016;7:102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Ionova-Martin SS, Wade JM, Tang S, et al. Changes in cortical bone response to high-fat diet from adolescence to adulthood in mice. Osteoporos Int. 2011;22(8):2283–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Lecka-Czernik B, Stechschulte LA, Czernik PJ, Dowling AR. High bone mass in adult mice with diet-induced obesity results from a combination of initial increase in bone mass followed by attenuation in bone formation; implications for high bone mass and decreased bone quality in obesity. Mol Cell Endocrinol. 2015;410:35–41. [DOI] [PubMed] [Google Scholar]
  • 22.Ehrich TH, Kenney-Hunt JP, Pletscher LS, Cheverud JM. Genetic variation and correlation of dietary response in an advanced intercross mouse line produced from two divergent growth lines. Genet Res. 2005;85(3):211–22. [DOI] [PubMed] [Google Scholar]
  • 23.Reich MS, Jarvis JP, Silva MJ, Cheverud JM. Genetic relationships between obesity and osteoporosis in LGXSM recombinant inbred mice. Genet Res (Camb). 2008;90(5):433–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Cheverud JM, Lawson HA, Bouckaert K, et al. Fine-mapping quantitative trait loci affecting murine external ear tissue regeneration in the LG/J by SM/J advanced intercross line. Heredity (Edinb). 2014;112(5):508–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.MacArthur J Genetics of body size and related characters. Selection of small and large races of the laboratory mouse. Am Nat. 1944;78:142–57. [Google Scholar]
  • 26.Goodale HD. A study of the inheritance of body weight in the albino mouse by selection. J Hered. 1938;29:101–12. [Google Scholar]
  • 27.Wilson SP, Goodale HD, Kyle WH, Godfrey EF. Long term selection for body weight in mice. J Hered. 1971;62(4):228–34. [DOI] [PubMed] [Google Scholar]
  • 28.Cheverud JM, Ehrich TH, Kenney JP, Pletscher LS, Semenkovich CF. Genetic evidence for discordance between obesity- and diabetes-related traits in the LGXSM recombinant inbred mouse strains. Diabetes. 2004;53(10):2700–8. [DOI] [PubMed] [Google Scholar]
  • 29.Brodt MD, Ellis CB, Silva MJ. Growing C57Bl/6 mice increase whole bone mechanical properties by increasing geometric and material properties. J Bone Miner Res. 1999;14(12):2159–66. [DOI] [PubMed] [Google Scholar]
  • 30.Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Muller R. Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res. 2010;25(7):1468–86. [DOI] [PubMed] [Google Scholar]
  • 31.Jepsen KJ, Silva MJ, Vashishth D, Guo XE, van der Meulen MC. Establishing biomechanical mechanisms in mouse models: practical guidelines for systematically evaluating phenotypic changes in the diaphyses of long bones. J Bone Miner Res. 2015;30(6):951–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Schriefer JL, Robling AG, Warden SJ, Fournier AJ, Mason JJ, Turner CH. A comparison of mechanical properties derived from multiple skeletal sites in mice. J Biomech. 2005;38(3):467–75. [DOI] [PubMed] [Google Scholar]
  • 33.Jepsen KJ, Akkus O, Majeska RJ, Nadeau JH. Hierarchical relationship between bone traits and mechanical properties in inbred mice. Mammalian Genome. 2003;14(2):97–104. [DOI] [PubMed] [Google Scholar]
  • 34.Silva MJ, Brodt MD, Fan Z, Rho JY. Nanoindentation and whole-bone bending estimates of material properties in bones from the senescence accelerated mouse SAMP6. J Biomech. 2004;37(11):1639–46. [DOI] [PubMed] [Google Scholar]
  • 35.van Lenthe GH, Voide R, Boyd SK, Muller R. Tissue modulus calculated from beam theory is biased by bone size and geometry: implications for the use of three-point bending tests to determine bone tissue modulus. Bone. 2008;43(4):717–23. [DOI] [PubMed] [Google Scholar]
  • 36.Tian L, Yu X. Fat, Sugar, and Bone Health: A Complex Relationship. Nutrients. 2017;9(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Choi HS, Park SJ, Lee ZH, Lim SK. The Effects of a High Fat Diet Containing Diacylglycerol on Bone in C57BL/6J Mice. Yonsei Med J. 2015;56(4):951–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Wu S, Aguilar AL, Ostrow V, De Luca F. Insulin resistance secondary to a high-fat diet stimulates longitudinal bone growth and growth plate chondrogenesis in mice. Endocrinology. 2011;152(2):468–75. [DOI] [PubMed] [Google Scholar]
  • 39.Behringer M, Gruetzner S, McCourt M, Mester J. Effects of weight-bearing activities on bone mineral content and density in children and adolescents: a meta-analysis. J Bone Miner Res. 2014;29(2):467–78. [DOI] [PubMed] [Google Scholar]
  • 40.Lawson HA, Lee A, Fawcett GL, et al. The importance of context to the genetic architecture of diabetes-related traits is revealed in a genome-wide scan of a LG/J x SM/J murine model. Mamm Genome. 2011;22(3–4):197–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Wang C, Tian L, Zhang K, et al. Interleukin-6 gene knockout antagonizes high-fat-induced trabecular bone loss. J Mol Endocrinol. 2016;57(3):161–70. [DOI] [PubMed] [Google Scholar]
  • 42.Zhang K, Wang C, Chen Y, et al. Preservation of high-fat diet-induced femoral trabecular bone loss through genetic target of TNF-alpha. Endocrine. 2015;50(1):239–49. [DOI] [PubMed] [Google Scholar]
  • 43.Upadhyay J, Farr OM, Mantzoros CS. The role of leptin in regulating bone metabolism. Metabolism. 2015;64(1):105–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.El-Haschimi K, Pierroz DD, Hileman SM, Bjorbaek C, Flier JS. Two defects contribute to hypothalamic leptin resistance in mice with diet-induced obesity. J Clin Invest. 2000;105(12):1827–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Lin S, Thomas TC, Storlien LH, Huang XF. Development of high fat diet-induced obesity and leptin resistance in C57Bl/6J mice. Int J Obes Relat Metab Disord. 2000;24(5):639–46. [DOI] [PubMed] [Google Scholar]
  • 46.(CGD) CfGD. Multi-system survey of mouse physiology in 72 inbred strains of mice (ANOVA-adjusted methodology) MPD:CGDpheno1. Mouse Phenome Database web resource (RRID:SCR_003212), The Jackson Laboratory, Bar Harbor, Maine USA: https://phenome.jax.org. [Google Scholar]
  • 47.Salem GH, Dennis JU, Krynitsky J, et al. SCORHE: a novel and practical approach to video monitoring of laboratory mice housed in vivarium cage racks. Behav Res Methods. 2015;47(1):235–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Clarke KA, Smart L, Still J. Ground reaction force and spatiotemporal measurements of the gait of the mouse. Behav Res Methods Instrum Comput. 2001;33(3):422–6. [DOI] [PubMed] [Google Scholar]
  • 49.Schmitt D, Zumwalt AC, Hamrick MW. The relationship between bone mechanical properties and ground reaction forces in normal and hypermuscular mice. J Exp Zool A Ecol Genet Physiol. 2010;313(6):339–51. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S3
NIHMS1053070-supplement-Table_S3.xlsx (18.5KB, xlsx)
supp figs
NIHMS1053070-supplement-supp_figs.pdf (719.3KB, pdf)
supp tabs
NIHMS1053070-supplement-supp_tabs.pdf (85.7KB, pdf)
suppl methods
NIHMS1053070-supplement-suppl_methods.docx (12.8KB, docx)

RESOURCES