Bone resorption is associated with the release of phosphate, in addition to calcium, from bone because of the dissolution of hydroxyapatite. Supportive of this notion that serum phosphate is partly originated from the release of bone is our recent reports [13] that cinacalcet administration significantly suppressed serum phosphate level in hemodialysis patients and our finding that administration of denosumab suppressed serum phosphate by approximately 10 % (unpublished results). Of great interest, our data demonstrated that groups of hemodialysis patients who exhibited higher serum active PTH(1-84) fraction [14] and bone alkaline phosphatase [15], and significant bone loss [16] showed significantly higher all-cause mortality than those without bone loss. Administration of cinacalcet to 5/6-nephrectomized hyperparathyroid rats protected against the development of vascular calcification at aortic arch [17]. In humans, aortic calcification is often observed in CKD patients, particularly in those with increased bone resorption resulting from renal hyperparathyroidism. Furthermore, Together with our data indicating that, even in non-CKD postmenopausal osteoporotic women, administration of risedronate attenuated age-related increase of pulse wave velocity (PWV), a clinically relevant marker for arterial wall stiffening [18], the data indicating the significant protective effect of bisphosphonate against the development of acute myocardial infarction in osteoporotic patients [19], the increased release of phosphate from bone is capable of directly damaging vascular endothelial cells and inducing vascular smooth muscle cells to dedifferentiate into osteoblasts to cause vascular calcification [20]. Our previous study [20] showed that increase of phosphate levels from 1.4 to 2.0 mM caused calcification of vascular muscle cells in vitro after 14 days of culture, and that phosphate entry inside the cell is the trigger to induce dedifferentiation of vascular muscle cells to osteoblasts to develop calcification. Furthermore, increased phosphate load into circulation is known to deteriorate renal function to decrease glomerular filtration rate and increase proteinuria/albuminuria [21]. This mechanism might be involved in the harmful effect of phosphate to accelerate vascular injury in DM patients.

Therefore, al least in the early stage of DM patients under poorly controlled conditions, it is suggested that poor glycemic control increases urinary loss of calcium, stimulate bone resorption by inducing secondary hyperparathyroidism and that the resultant increased phosphate release from bone causes vascular damage, directly and indirectly by kidney damage, in DM patients.

Sustained 7-day exposure to high glucose significantly inhibited growth of human osteoblast-like MG-63 cells in a dose-dependent manner, in contrast with the insignificant suppression of high mannitol, osmolality control [10]. The mechanism for the suppression of osteoblasts by high glucose is explained by the direct suppressive effect of high glucose on cell proliferetaion through an intracellular accumulation of soribitol [10] and its indirect effect to attenuate IGF-1 stimulation of cell growth. In agreement with these data, the number of osteoblasts/osteocyte is reported to decrease significantly, resulting in the development of ABD in DM patients. Osteoblasts, in response to PTH and 1,25-dihydroxyvitamin D, an active form of vitamin D, increase intracellular cytosolic Ca⁺⁺ and osteocalcin production/secretion in vitro. Sustained 7-day exposure to high glucose significantly impaired the responsiveness of human osteoblast-like MG-63 cells to these hormones in a dose-dependent manner, in contrast with the insignificant suppression of high mannitol, osmolality control [22, 23]. These data may suggest that high glucose condition and/or insulin/IGF-1 deficiency impaired the proliferation and the cell responsiveness of osteoblast/osteocyte to cause ABD in DM patients.

In vitro system, calcium level in culture medium, independent of phosphate level, stimulates calcification of vascular smooth muscle cells in a dose-dependent manner, suggesting the direct effect of calcium on vascular calcification. The enhancement of aortic calcification by increased calcium load is also evidenced in vivo in hemodialysis patients, although the aortic calcification score was significantly affected by bone turnover status [8]. Since administration of calcium-containing phosphate binder to CKD patients is known to induce vascular calcification significantly more than non-calcium containing phosphate binder [24]. Although the direct effect of calcium load to enhance vascular calcification is hypothesized on the basis of in vitro study, the main mechanism by which calcium load to stimulate vascular calcification in human subjects is explained by its indirect effect to suppress bone turnover by suppressing parathyroid function [8, 25]. As bone turnover status becomes slower, the rate of bone calcification could be progressed too high to adsorb more calcium/phosphate from circulation. This phenomenon could easily elevate serum calcium/phosphate to cause ectopic calcification including vascular wall. We recently reported that replacement of calcium carbonate (CaC), a calcium-containing phosphate binder, with lanthanum carbonate (LaC), a non-calcium containing phosphate binder, increased serum PTH and bone turnover markers in hemodialysis patients with suppressed serum PTH≦150 pg/mL [26], along with the re-appearance of double tetracycline labeling on bone biopsy specimens obtained from such hemodialysis patients [27]. Therefore, it is suggested that LaC normalized CaC-induced suppression of parathyroid function and bone turnover by decreasing calcium overloading.