Interaction of sexual dimorphism and gene dosage imbalance in skeletal deficits associated with Down syndrome

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

J o u r n a l P r e -p r o o f measurement of 20mm. Femur (measured from proximal epiphysis to distal epiphysis) and rostrocaudal (tip of the soft tissue of the nose to base of tail) lengths were measured via ImageJ.
The femur was analyzed due to its prevalence in the Jackson Laboratory Phenome database and previous work in DS mouse models.
Trabecular analysis was performed on the distal metaphysis with a region of interest (ROI) defined as 10% of the total bone length, approximately 1mm proximal to the distal growth plate and then extending proximally. The ROI was auto-segmented using a custom Matlab (MathWorks, Inc. Natick, MA) code [45]. Measurements of trabecular architecture (bone volume fraction [BV/TV], trabecular thickness, number, and separation), bone mineral density (BMD), and tissue mineral density (TMD) were calculated using CTAn.
Cortical analysis was performed at a standard site 60% of the total bone length away from the distal growth plate. Seven transverse slices were generated from the standard site and J o u r n a l P r e -p r o o f 11 cortical geometric properties were obtained from using a custom Matlab code.

Mechanical testing
The mechanical properties of the femur were determined by 3-point bending using a mechanical testing machine (TA ElectroForce 3200; Eden Prairie, MN, USA). The femurs were thawed to room temperature and tested in the anterior-posterior direction with the posterior surface in compression (7 mm support span). The load was applied to the midpoint of each bone.
The femur was preloaded using 0.5N to establish contact with the bone and then monotonically tested to failure at a displacement rate of 0.025 mm/s while fully hydrated. Load and deflection were recorded, from which structural strength (yield and ultimate), stiffness (slope of the linear portion of the force versus displacement curve), and deformation (yield deformation, postyield deformation and total deformation) were determined [46,47]. Distance from the proximal end of the femur to the location of the fracture initiation was measured using calipers. Seven transverse slices were obtained from µCT at the location of fracture and calculated geometric properties (bending moment of inertia and distance from the centroid to the tensile surface of the bone).
Along with the deflection data, the moment of inertia and distance from the centroid to the tensile surface of the bone in tension derived were used to map load-displacement into stress vs. strain curves using standard equations derived from Euler-Bernoulli beam theory to estimate tissue level properties [46]. The mechanical strength, stiffness, and work/toughness were determined from the force vs. displacement and stress vs. strain curve. In these equations, F is the force, d is the displacement, a is the distance from the support to the inner loading point (4 mm), and L is the span between the outer supports (7 mm). The yield point was calculated using the 0.2% offset method based on the stress-strain curve. The modulus of elasticity was calculated as the slope of the linear portion of the stress-strain curve.

Histomorphometric analyses
Male and female mice were injected intraperitoneally with 0.6% Calcein green diluted in PBS at 15 weeks of age, and injected again 4 days later. Three days after the second injection of Calcein green, the mice were euthanized and weighed (16 weeks of age). Femurs were removed and placed in 70% ethanol and stored at room temperature until ready for use. The femur was separated from at the midshaft and the proximal and distal femurs were processed, cut, and sectioned as described [36]. Briefly, femurs were dehydrated in graded levels of ethanol, cleared in xylene and embedded in methyl methacrylate. For dynamic histology, the midshaft of the femur was sectioned into 500µm transverse sections and ground to 40µm and then mounted using Eukitt to enhance viewing of fluorescent label. One section was read using a D-FL Epi-

Statistical Tests
MicroCT and mechanical testing data were analyzed using a custom Matlab code [45].
For microCT, mechanical testing, and histomorphometry data, normality and homogeneity were assessed for each phenotype and any violations were transformed to achieve normality.

Interaction of sex and genotype in Dp1Tyb mice alters BV/TV during skeletal maturation
The interaction of sex (male vs. female) and genotype (three copy vs. normal copy number) was important in percent bone volume (bone volume/tissue volume or BV/TV) at both 6 and 16 weeks of age. At 6 weeks of age during longitudinal bone growth, BV/TV was reduced in male Dp1Tyb, female Dp1Tyb and female control animals as compared to male control animals ( Fig. 1A). At 16 weeks of age, a time of skeletal maturity, BV/TV continued to be reduced in male Dp1Tyb, female Dp1Tyb and female control animals as compared to male control animals.
Additionally, female Dp1Tyb mice had reduced BV/TV as compared to male Dp1Tyb mice (Fig.   1A). BV/TV increased in male mice from 6 to 16 weeks of age (p<0.001), but female mice exhibited a loss in BV/TV during the same time period (p=0.025). Similar data were observed for BMD in the trabecular compartment at both 6 and 16 weeks. Taken together, these data J o u r n a l P r e -p r o o f 16 indicate that percent bone volume is reduced in Dp1Tyb as compared to wild-type male mice as the bone is actively growing, and approximately when skeletal maturity is achieved. Female mice have reduced formation or bone accrual independent of genotype at 6 weeks. At 16 weeks of age, as the appendicular skeleton reaches skeletal maturity, BV/TV is increased in male mice, but decreased in female mice as compared to 6 weeks of age.
3.3 Interaction between sex and genotype affects trabecular microarchitecture in male and female Dp1Tyb and wild-type mice Measurements of other trabecular skeletal parameters provide insight into the interaction between sex and genotype. Trabecular number (Tb.N) is increased in male wild-type mice compared to female Dp1Tyb and wild-type mice at 6 and 16 weeks (Fig. 1B) and trabecular thickness (Tb.Th) is greater in male wild-type mice than male and female Dp1Tyb and wild-type mice at both 6 and 16 weeks (Fig. 1C). Trabecular separation (Tb.Sp) is greater in female than male mice at 16 weeks (Fig. 1D) J o u r n a l P r e -p r o o f 3.4 Cortical bone parameters exhibit differences between sex and genotype in Dp1Tyb and control mice at 6 weeks, and interactive effects at 16 weeks Six week-old control male and female mice had a greater total cross-sectional area (total CSA) in the cortical bone than Dp1Tyb male and female mice ( Fig. 2A) 3). Also at 16 weeks, bone from Dp1Tyb as compared to control mice displayed a higher yield stress (normalized force to size and shape of bone), ultimate stress, and resilience. Control as compared to Dp1Tyb mice had a higher total strain. Female mice had marginally increased yield stress as compared to male mice. Taken together, these data indicate that at 16 weeks, Dp1Tyb and female mice seem to have improved material bone properties as compared to control and male mice, especially in the elastic region before the bone becomes permanently deformed. In the trabecular bone, there was an increase in female control as compared to male control bone in MS/BS (p=0.02) and BFR (p=0.02) (sex × genotype interaction) ( Fig. 5A-C).
Bone from female as compared to male mice exhibited a higher trabecular MAR (p=0.02) at 16 weeks of age (Fig. 5C). Osteoid surface per bone surface (OS/BS) (p=0.04) and OS/BS percentage (p<0.001) were also greater in trabecular bone of female as compared to male mice.
Male Dp1Tyb mice had a lower osteoclast number on the bone surface (Oc#/BS) as compared to all other mice at 16 weeks of age (sex × genotype interaction, p=0.02) (Fig 5D) as measured by Tartrate-resistant acid phosphatase (TRAP) labeled multinucleated osteoclasts. Additionally, the percentage of bone surface covered by osteoclasts (OcS/BS) was significantly reduced in male as compared to female (p=0.004) and control as compared to Dp1Tyb (p=0.0001) mice (Fig 5D).

Discussion
The gives insight into some parameters, including BMD, but does not provide information on the geometry or composition of bone. BMD is a metric used to calculate risk factor for fracture, however, many studies do not include geometry, structural and strength indices that also affect fracture incidence [37,48].

Similarities between humans with DS and DS mouse models
It is hypothesized that the osteoporotic phenotype found in individuals with DS is highly influenced by limited bone mineral accretion and reduced peak bone mass attainment during adolescence [11,30]. In humans the age of total body peak bone mass has been estimated to be about 18.8 years in females and 20.5 years in males [49]. Individuals with DS attain peak bone mass earlier and at lower levels than normal individuals [11]. Women with DS (as well as normal women) have bone mineral accrual later than men, but experience rapid bone loss after the age of 40 [30]. Men with DS have a gradual loss of bone after early adulthood, similar to men without DS, but the rate of loss is increased in men with DS [14,30].
Information from the analyses described herein concentrated on skeletal properties during bone accrual (6 weeks) until the estimated age of skeletal maturity in mice (16 weeks) [50].
There are few studies that have examined DS skeletal deficits in humans during adolescence and J o u r n a l P r e -p r o o f 24 the time of peak bone accrual, and the results from Dp1Tyb male and female mice may help explain previous reports. The lower BMD seen in adolescent males and females with DS analyzed together as compared to normal individuals [25,26] compares to the genotype effect causing low BV/TV (and BMD) in 6-week Dp1Tyb mice. A study of adolescents with DS that found no differences in BMD between males and females [27] correlates to the similar BV/TV levels in 6 week old male and female Dp1Tyb mice. At an age corresponding to skeletal maturity in individuals with DS, both males and females had a lower BMD [1], with DS females having a lower BMD than DS males in their limbs [2]; both male and female Dp1Tyb mice had significantly affected bone structure at 16 weeks, and trabecular and cortical values were significantly lower in females than males at 16 weeks. The data presented herein also demonstrate that male Dp1Tyb mice exhibit osteopenic phenotypes earlier than their control counterparts, similar to what has been observed in individuals with DS [14,28] and that osteoporotic phenotypes affect males as well as females in early adult stages [30].

Comparison of Dp1Tyb and other DS model mice
Bone deficits associated with DS were first identified in Ts65Dn male mice [36]. A comparison between these and previously published results reveals similar direction and magnitude in trabecular, cortical and mechanical deficits in mice that are at dosage imbalance for genes homologous to Hsa21 [36,39]. Previous studies of 6, 12, 16 week, and 24 month old male mice found no significant differences in trabecular thickness (Tb.Th) between Ts65Dn and euploid mice, though these values were near a p<0.05 significance level [36,37]. Tb.Th was significantly different between Ts65Dn and control mice at 6 weeks of age and were corrected in Ts65Dn,Dyrk1a+/-mice in a subsequent study [39]. Ts1Rhr as compared to Dp1Tyb mice could dilute any potential effect from triplicated genes.
Differences between skeletal abnormalities could also be due to genetic background, number and which genes are at dosage imbalance, and origin of the change in genetic dosage.
Differences in skeletal phenotypes could be due to differences in triplicated gene content or  Ts65Dn as compared to control mice showed reduced osteoblast number and/or activity coupled with increased osteoclast activity caused the DS like skeletal deficits at that age [36,39]. A low bone turnover hypothesis was proposed for the deficits skeletal deficits seen in 12-week-old male Ts65Dn mice because of decreased osteoclast and osteoblast activity found on proximal tibia and femur as compared to euploid mice at this age [37]. In 16-week-old male mice, we observed increased osteoblast activity and reduced osteoclast number and activity in the trabecular compartment of Dp1Tyb as compared to control mice. In female mice of the same age, there was a decrease in Dp1Tyb osteoblast activity. These results indicate that osteoblast and osteoclast activity was different in 16-week-old male Dp1Tyb mice than previously reported osteoblast and osteoclast activity in Ts65Dn mice at 6 and 12 weeks. Only a slight increase in osteoblast activity on the periosteal surface of male Dp1Tyb cortical bone was observed. Taken together, these data suggest that sex, gene dosage, bone location and age are all important in the cellular components Dp1Tyb as compared to control mice. Furthermore, the cellular mechanisms leading to bone deficits associated with DS may be distinct at different points in development.
These data demonstrate fundamental differences in skeletal development between