While the rat model of DM in the present study demonstrated bone fragility on mechanical testing, µCT analysis revealed that the cortical bone mineral density had actually increased. Bone histomorphometry revealed a decrease in osteocyte count and an increase in mineral deposit within the cortical bone of the harvested right tibiae. To our knowledge, this is the first study to demonstrate increased mineral deposit in the cortical bone of diabetic rats.
The WBN/Kob rat is a model of DM resulting from pancreatic exocrine hypofunction, which has been used in the past to assess cortical bone (Mori et al., 1992). We considered that this model was valid in our study, exhibiting a significant elevation in the level of blood glucose and GA. We confirmed that neither blood urea nitrogen nor creatinine levels were elevated, indicating that renal function was not diminished and, as such, identified changes in the bones in our study were not dependent on diminished renal function. In DM, insulin deficiency causes inhibition of osteoblast activity, which can reduce (or even inhibit) cortical bone modeling. This could lead to reductions in the overall bone size and numbers of osteocytes (Bouillon et al., 1995).
Bone mineral density, measured by µCT, was increased in the diabetic rats. Considering the fact that cortical porosity was significantly lower in the diabetic rats compared with that in the control rats, according to histomorphometric evaluation, this may have resulted in the relative increase in bone mineral density. A number of reports have also shown that bone mineral density increased in patients with DM (Schwartz et al., 2001; Ma et al., 2012). Subsequently, cortical porosity may be relatively decreased, which is consistent with the present results.
On mechanical testing, the tibiae in the diabetic group sustained smaller maximum loads and reduced bone energy absorption than the control group, but without a lower stiffness. Thus, the tibiae in the diabetic rats were hard, but weak. This result is consistent with previous studies that reported an increase in abnormal collagen crosslinking associated with DM (Saito et al., 1997). Accumulation of advanced glycosylation end products increases abnormal collagen crosslinking, which in turn results in large numbers of bonds between collagen molecules. This phenomenon may have reduced collagen energy absorption and, therefore, bone energy absorption in the subjects in this study.
Bone histomorphometry revealed a significantly greater volume of mineral deposit in the diabetic group compared with in the control group. In general, calcification is broadly divided into the following processes: mineralization, which is driven by osteoblasts; endochondral ossification, which is driven by chondrocytes; and intramembranous ossification, which is driven by mesenchymal cells (Eriksen et al., 1994). The mineral deposit observed in the diabetic group in our study was considered not to fall under any of the above classifications. Specifically, mineral deposit was significantly greater in the diabetic group than in the control group, creating the appearance that bone mineral density was preserved. The number of osteocytes was decreased in the diabetic group compared with that in the control group, which may be due to the apoptosis of osteocytes by secondary mineralization. It is speculated that the decrease in the number of osteocytes may have reduced the amount of matrix, such as collagen, which should have been produced by the osteocytes. In fact, this mineral deposit was abnormal mineralization, possibly resulting in reduced bone absorption and bone strength. Hunt et al. reported that mineral crystallinity was observed in patients with diabetes by FTIR (Hunt et al., 2021). This mineral crystallinity may be related to the mineral deposit found in this study. Thus, abnormal mineral deposit may be a factor in the reduced bone strength associated with DM.
A strength of our study was the use of an animal model of DM, which allowed us to conduct invasive mechanical testing and histological assessments. Specifically, because we were able to obtain a whole bone specimen, we were able to prove bone fragility by mechanical testing, which has never been reported before, and to examine the relationship between this and histological evaluation in the same specimen. The limitations of our study must also be acknowledged. Specifically, identified cortical bone changes in our animal models may not occur in humans with DM. In addition, the use of individual samples limits the interpretation between histological changes and mechanical testing results.