Genetic analysis of triplicated genes affecting sex-specific skeletal deficits in Down syndrome model mice

Down syndrome (DS) is caused by the triplication of human chromosome 21 (Hsa21), resulting in skeletal insufficiency (low bone mineral density) and altered bone development. DS mouse models recapitulate these deficits, including sexual dimorphism in long bone alterations. Historically, Ts65Dn mice provided much of the insight behind DS-related skeletal deficits with ~100 trisomic orthologous genes, but there are concerns about the genetic fidelity in this model due to the included triplication of genes not homologous to Hsa21. A new DS model, Ts66Yah, subtracted the non-Hsa21 homologous trisomic genes from Ts65Dn but has not been evaluated for long bone deficits. Comparing skeletal phenotypes between these models can determine the contribution of non-Hsa21 homologous trisomic genes and whether the Ts66Yah mouse is relevant as a model for DS-associated skeletal deficits. After assessing individual densitometric, morphometric, and mechanical variables in male and female Ts66Yah femurs at similar ages to when skeletal deficits were observed in Ts65Dn mice, structural phenotypes were directly compared to those of Ts65Dn mice using a novel multivariate principal components analysis method to generate composite scores. Overall, structural and mechanical bone phenotypes of the femur appeared milder in Ts66Yah compared to Ts65Dn mice. The appearance of developmental trabecular microarchitecture deficits, but not other abnormalities, were evident earlier in Ts65Dn than Ts66Yah mice. Dyrk1a, a gene triplicated in both models, affected skeletal structure differently in each model, likely through differing gene interactions. The novel principal components analysis detected subclinical phenotypes lost in individual analyses, which could be advantageous when determining overall skeletal deficits.

Methods
Six-, nine-, and sixteen-week-old Ts66Yah femurs were wrapped in parafilm to maintain hydration and single-scanned using a SkyScan 1172 µCT system (Bruker, Kontich, Belgium) with the following parameters: 60kV, 167uA, 885ms, 10-micron voxel size, Al 0.5mm filter, 0.7⁰ rotation step and frame averaging of two. Two hydroxyapatite phantoms (0.25 and 0.75g/cm3 CaHA) were used to calibrate bone mineral density for each scanning session.

Femurs from P36 mice were wrapped in parafilm and group-scanned (Kohler et al. 2021) using a SkyScan 1272 µCT system (Bruker, Kontich, Belgium) with the following parameters: 70kV, 142uA, 1265ms, 10-micron voxel size, Al 0.5mm filter, 0.7⁰ rotation step and frame averaging of two. Two hydroxyapatite phantoms (0.3 and 1.25g/cm3 CaHA) were used to calibrate bone mineral density for each scanning session. No unstandardized comparisons were made between scans from different µCT systems.

All scans were reconstructed using NRecon with the following settings: ring artifact reduction 5, beam-hardening correction 20%, dynamic image (attenuation coefficient) boundaries 0.00 and 0.11. Reconstructed scans were rotated so the anterior surface of the distal growth plate faces right in the transaxial view for all bones. Using CTAnalyzer (CTAn), the trabecular region of interest (ROI) was defined as a 1mm section extending proximally from the end of the distal growth plate and isolated from the cortical bone. The cortical ROI was defined as a 1mm (6-, 9-, and 16-week-old mice) or 0.65mm (P36 mice) section extending distally from the beginning of the third trochanter. Trabecular bone parameters, including bone mineral density (BMD), bone volume fraction (BV/TV), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), and trabecular number (Tb.N), were calculated using internal CTAn functions. Cortical bone parameters, including total cross-sectional area (Tt.Ar), marrow area (Ma.Ar), cortical bone area (Ct.Ar), cortical bone area fraction (Ct.Ar/Tt.Ar), cortical thickness (Ct.Th), periosteal perimeter (Ps.Pm), endocortical perimeter (Ec.Pm), maximum and minimum moment of inertia (Imax and Imin) for P36 bones, and cortical tissue mineral density (Ct.TMD), were calculated using a custom MATLAB code (Berman et al. 2015; Stringer et al. 2017).

Mechanical properties were determined as described previously (Thomas et al. 2020). Briefly, 3-point femur bend was performed after µCT analysis of 6-, 9-, and 16-week-old right femurs using a TA ElectroForce 5500 testing machine while bones were hydrated using PBS (Eden Prairie, MN, USA). Right femurs were tested using a 5-, 6-, or 7mm support span (6, 9, and 16 weeks, respectively) in the anterior-posterior direction with the posterior surface in compression. A 10lb load was utilized for 6- and 9-week-old femurs, and a 50lb load was utilized for 16-week-old femurs. All bones were preloaded (0.2-0.4 N) to establish contact with the loading point located at the midshaft, then testing occurred at 0.025mm/sec to failure. The 0.2% offset method was utilized on the slope of the linear portion of the stress-strain curve to find the yield point. The ultimate point was determined as the maximum force reached, while the failure point was determined as where the bone broke. Two 6-week-old female euploid femurs were excluded from analysis due to not reaching a failure point before maximum displacement at 3.5mm. The following whole bone (extrinsic) parameters were reported from the force-displacement curve: yield and ultimate force, displacement and work to yield, postyield displacement and work, total displacement and work, and stiffness. Cortical geometry from µCT analysis was utilized to normalize the stress-strain curve from the force-displacement curve. The following tissue-estimate (intrinsic) properties were reported from the stress-strain curve: yield and ultimate stress, strain to yield, total strain, modulus, resilience, and toughness.

References

Berman AG, Clauser CA, Wunderlin C, Hammond MA, Wallace JM. 2015. Structural and mechanical improvements to bone are strain dependent with axial compression of the tibia in female c57bl/6 mice. PLoS One. 10(6):e0130504.

Kohler R, Tastad CA, Stacy AJ, Swallow EA, Metzger CE, Allen MR, Wallace JM. 2021. The effect of single versus group muct on the detection of trabecular and cortical disease phenotypes in mouse bones. JBMR Plus. 5(4):e10473.

Stringer M, Abeysekera I, Thomas J, LaCombe J, Stancombe K, Stewart RJ, Dria KJ, Wallace JM, Goodlett CR, Roper RJ. 2017. Epigallocatechin-3-gallate (egcg) consumption in the ts65dn model of down syndrome fails to improve behavioral deficits and is detrimental to skeletal phenotypes. Physiol Behav. 177:230-241.

Thomas JR, LaCombe J, Long R, Lana-Elola E, Watson-Scales S, Wallace JM, Fisher EMC, Tybulewicz VLJ, Roper RJ. 2020. Interaction of sexual dimorphism and gene dosage imbalance in skeletal deficits associated with down syndrome. Bone.115367.