Although we focus on characterization of human tissue in translational studies that elucidate microstructural insights that can directly be related to patient care, animal models of skeletal disease are critical for preclinical assessment of disease mechanisms and drug therapies. In particular, based on our history of detailed materials characterization with FTIR imaging and Raman imaging, we are leaders in vibrational spectroscopic imaging characterization of disease- and drug-induced changes in bone tissue properties. Our sophisticated measurement techniques, which enable sensitive, spatially resolved vibrational spectroscopic measurements, and our detailed data analysis techniques, for which we have recently developed more powerful statistical models and improved validations, distinguish our work in this area. 

We have interrogated many different systems with these tools. Some representative examples include (a) mouse models of type 2 diabetes [Hunt et al. 2018] ; (b) ovariectomized rats treated with several anti-osteoporotic agents  [Taylor et al. 2020]; (c) mouse models of altered gut microbiome (with Prof. Christopher Hernandez) [Guss et al 2020]; and (d) pathologic mineralization in aortic valves (with Profs. Lara Estroff and Jonathan Butcher) [Richards et al. 2018]

3.1.1 Effects of sustained hyperglycemia on bone material properties in a mouse model of type 2 diabetes

We complement our work in human type 2 diabetes with work in murine models. In this work, we examined bone tissue properties in the KK-Ay murine model of T2DM. PhD student Heather Hunt assessed the effects of prolonged hyperglycemia on bone tissue compositional properties, enzymatic collagen crosslinks, and nonenzymatic collagen crosslinks called advanced glycation endproducts (AGEs) in KK-Ay mice, using FTIR imaging and HPLC. This work was performed in collaboration with Prof. Karen King at University of Colorado Medical School.

Key findings We showed that  bone tissue of KK-Ay mice (which mimic type 2 diabetes) exhibited increased collagen maturity, increased mineral content, and less heterogeneous mineral properties relative to KK-aa littermate controls, which mimic the non-diabetes state [Hunt et al. 2018].

Representative FTIR images a/a (control) and Ay/a (type 2 diabetic) mouse bone for (A) cortical collagen maturity (XLR), (B) whole proximal femur (WPF) collagen maturity, (C) trabecular collagen maturity, and (D) whole proximal femur mineral:matrix ratio. Scale bars: 50 m in A,C and 500m in B,D. [Hunt et al. 2018].

Impact These measurements are the first assessment of skeletal compositional properties in the KK-Ay mouse model. The observed compositional differences in the KK-Ay mouse tissue are consistent with older bone in the KK-Ay mice (greater time since formation). Most importantly, we have shown that observations that humans with type 2 diabetes exhibit reduced bone remodeling are also seen in our mouse model of the same disease.

Future work In next steps, we will determine the functional consequences of the observed differences in compositional properties through multiscale mechanical and microstructural assessment of the bones of the KK-Ay mice vs littermate controls.

3.1.2 Sequential therapies for treatment of osteoporosis

Antiresorptive and anabolic agents are used sequentially to treat osteoporosis, but the effects of sequential treatment on bone material properties are incompletely understood. Therefore, in this study, PhD student Erik Taylor compared cortical bone material composition from a model of ovariectomized rats (the standard preclinical rodent model of postmenopausal osteoporosis) treated with multiple sequential treatment regimens of three antiresorptive agents (alendronate, raloxifene, and risedronate) and one anabolic agent (h-PTH 1-34). This work was performed in collaboration with Dr. Nancy Lane at UC Davis.

Key findings We showed that sequential treatments that included the anabolic agent PTH had lower mineral content relative to the vehicle control and lower collagen maturity relative to the monotherapy control [Taylor et al. 2020] .

Impact The observed changes in the mineral and organic phases suggest that sequential use of anti-resorptive and anabolic treatments promote and maintain new bone formation, as inferred from improved cortical bone tissue properties relative to monotherapies.  This means that monotherapies, the current standard clinical implementation for most patients with postmenopausal osteoporosis, may be improved with sequential therapies.

Future work Our goal is to extend these studies to humans to determine if these results can be translated to clinical care.

Karen King, University of Colorado School of Medicine
Nancy Lane, UC Davis
Christopher Hernandez, Cornell University
Lara Estroff, Cornell University
Jonathan Butcher, Cornell University