The vitamin D receptor in osteoblastic cells but not secreted parathyroid hormone is crucial for soft tissue calcification induced by the proresorptive activity of 1,25(OH)2D3

The vitamin D receptor (VDR) is expressed most abundantly in osteoblasts and osteocytes (osteoblastic cells) in bone tissues and regulates bone resorption and calcium (Ca) and phosphate (P) homeostasis in association with parathyroid hormone (PTH). We previously reported that near-physiological doses of vitamin D compounds suppressed bone resorption through VDR in osteoblastic cells. We also found that supra-physiological doses of 1α,25-dihydroxyvitamin D3 [1,25(OH)2D3] induced bone resorption and hypercalcemia via VDR in osteoblastic cells. Here, we report that the latter, a proresorptive dose of 1,25(OH)2D3, causes soft tissue calcification through VDR in osteoblastic cells. High concentrations of vitamin D affect multiple organs and cause soft tissue calcification, with increases in bone resorption and serum Ca levels. Such a variety of symptoms is known as hypervitaminosis D, which is caused by not only high doses of vitamin D but also impaired vitamin D metabolism and diseases that produce 1,25(OH)2D3 ectopically. To clarify the biological process hierarchy in hypervitaminosis D, a proresorptive dose of 1,25(OH)2D3 was administered to wild-type mice in which bone resorption had been suppressed by neutralizing anti-receptor activator of NF-κB ligand (RANKL) antibody. 1,25(OH)2D3 upregulated the serum Ca x P product, concomitantly induced calcification of the aorta, lungs, and kidneys, and downregulated serum PTH levels in control IgG-pretreated wild-type mice. Pretreatment of wild-type mice with anti-RANKL antibody did not affect the down-regulation of PTH levels by 1,25(OH)2D3, but inhibited the increase of the serum Ca x P product and soft tissue calcification induced by 1,25(OH)2D3. Consistent with the effects of anti-RANKL antibody, VDR ablation in osteoblastic cells also did not affect the down-regulation of PTH levels by 1,25(OH)2D3, but suppressed the 1,25(OH)2D3-induced increase of the serum Ca x P product and calcification of soft tissues. Taken together with our previous results, these findings suggest that bone resorption induced by VDR signaling in osteoblastic cells is critical for the pathogenesis of hypervitaminosis D, but PTH is not involved in hypervitaminosis D.


Introduction
The major role of the active form of vitamin D 3 , 1α,25-dihydroxyvitamin D 3 [1,25(OH) 2 D 3 ], is to maintain the serum concentration of calcium (Ca) within a normal range by enhancing intestinal absorption of Ca from the diet. When dietary Ca intake is insufficient, parathyroid hormone (PTH) and PTH-induced 1,25(OH) 2 D 3 act on osteoblasts and osteocytes (osteoblastic cells) to enhance the expression of receptor activator of nuclear factor-κB ligand (RANKL), a cytokine essential for the differentiation and function of osteoclasts [1][2][3][4].
Osteoblastic cells exhibit the highest expression of the vitamin D receptor (VDR) in bone tissues and respond to 1,25(OH) 2 D 3 [5][6][7]. Our previous study showed that by VDR in osteoblastic cells, supra-physiological concentrations of 1,25(OH) 2 D 3 administration to mice induced RANKL expression, osteoclastic bone resorption, hypercalcemia, and synthesis of fibroblast growth factor 23 (FGF23) [8]. FGF23 is an osteocyte-derived hormone that suppresses phosphate (P) reabsorption and increases Ca reabsorption in the kidneys [9,10]. Osteoblastic cells also express the PTH 1 receptor and, upon binding to PTH, upregulate RANKL expression to induce bone resorption [11,12]. Moreover, many studies have shown that intermittent administration of PTH to humans and mice promotes bone formation [13,14]. These differential effects of PTH are considered to be due to conformational states of the PTH 1 receptor, which affect downstream cascades [14,15]. The RANKL-induced osteoclasts then mobilize Ca and P from the bone. Thus, vitamin D and PTH maintain the serum concentration of Ca at supersaturation levels for deposition of hydroxyapatite crystals in the bone matrix [1][2][3][4]. An imbalance of bone resorption and formation caused by impaired vitamin D and PTH signaling leads to several pathological conditions such as rickets, osteomalacia, osteosclerosis, and osteoporosis [1][2][3][4]16]. Furthermore, bone turnover balance favoring bone resorption has been proposed to cause systemic imbalance of Ca and P, and leads to vascular calcification [17,18].
The proresorptive activity of 1,25(OH) 2 D 3 was detected only with high doses in vivo, and in vitro culture in the presence of osteoblastic cells [6]. We previously showed that high-dose administration of 1,25 (OH) 2 D 3 to wild-type mice stimulated osteoclastic bone resorption with increasing serum Ca, and intact FGF23 levels, and reducing body weight due to dehydration, all of which are typical symptoms of vitamin D toxicity, hypervitaminosis D [8,19]. Hypervitaminosis D is caused by not only high doses of vitamin D but also impaired vitamin D metabolism or coincident disease, such as granulomatous diseases and lymphoma, that produce 1,25(OH) 2 D 3 ectopically [19]. These proresorptive effects were not evident in neutralizing anti-RANKL antibody (clone OYC1, [20])-pretreated wild-type mice or Osterix (Osx)-VDR-conditional knockout (cKO) mice, in which VDR is ablated specifically in osteoblastic cells [8]. In contrast, near-physiological doses of 1,25(OH) 2 D 3 and its analogs suppressed bone resorption in vivo [5,21,22]. It was difficult to determine whether the up-and down-regulation of bone resorption by vitamin D in vivo is caused by direct actions on bone or as a consequence of mineral control by VDR signaling in extra-skeletal tissues. To clarify the role of VDR in bone cells for bone resorption, several research groups including ours conducted conditional deletion of the VDR gene in osteoblast precursors [5], osteoblasts [23], osteocytes [24], chondrocytes [25], osteoclast precursors [26], and osteoclasts [5,27,28] in mice. We showed that VDR in osteoblastic cells was essential for the proresorptive properties of high doses of 1,25(OH) 2 D 3 [8]. Moreover, we demonstrated that VDR in osteoblastic cells was involved in the antiresorptive properties of near-physiological doses of vitamin D compounds in vivo [5][6][7]. In contrast, the 1,25(OH) 2 D 3 -VDR system in chondrocytes or osteoclast-lineage cells is unlikely to be a major regulator of osteoclastogenesis under physiological and pathological conditions [25][26][27][28][29].
A recent study showed that osteocytic production of 1,25(OH) 2 D 3 was augmented in a mouse chronic kidney disease (CKD) model and contributed to the prevention of soft tissue calcification [30]. This locally synthesized 1,25(OH) 2 D 3 was shown to decrease bone morphogenetic protein-2 (BMP2) production and stimulate sclerostin (an inhibitor of bone formation) production in osteocytes. 1,25 (OH) 2 D 3 -induced sclerostin inhibited BMP2-induced transdifferentiation to osteoblast-like cells in the arterial wall [30].
In the present study, to further investigate the biological process hierarchy in hypervitaminosis D, we examined: (i) whether 1,25 (OH) 2 D 3 -induced bone resorption and the 1,25(OH) 2 D 3 -VDR system in osteoblastic cells are involved in the regulation of PTH levels, and (ii) whether they are also involved in 1,25(OH) 2 D 3 -induced soft tissue calcification. Theoretically, inhibition of bone resorption should lower serum Ca and consequently raise PTH. Although inhibition of bone resorption prevented 1,25(OH) 2 D 3 -induced soft tissue calcification, it did not affect serum PTH levels. This finding indicates that PTH secretion is not involved in hypervitaminosis D. The present study uncovers the critical role of bone resorption regulated by osteoblastexpressed VDR in soft tissue calcification associated with hypervitaminosis D.

Mice
Male C57BL/6 mice were purchased from Japan SLC (Shizuoka, Japan). Osx-Cre Tg/0 (C57BL/6 genetic background) were purchased from the Jackson Laboratory (Bar Harbor, ME) [31]. VDR-floxed mice (C57BL/6 genetic background) were generated as previously described [23] and kindly supplied by Dr. Shigeaki Kato. Mice were fed a standard diet containing 0.8 % Ca and 0.65 % phosphorus (P), and 2.3 IU/g vitamin D 3 (Hi-Durability IRRD M/R, LabDiet, St. Louis, MO). All mice were housed in a specific pathogen-free facility at Matsumoto Dental University at 24 ± 2 o C and 50-60 % humidity with a 12-hour light/dark cycle, and they were provided with sterilized water and food ad libitum. All animal studies were reviewed and approved by the Research Animal Care and Use Committee of Matsumoto Dental University.

Measurement of PTH
Serum concentrations of PTH were measured using a mouse PTH 1-84 ELISA kit (Quidel, San Diego, CA) and an iMark microplate reader (Bio-Rad, Hercules, CA).

Alizarin red staining
For analysis of calcification, the mice were anesthetized with isoflurane using a vaporizer (DS Pharma Biomedical, Osaka, Japan), and then subjected to blood collection via heart puncture and transcardial perfusion with 4 % paraformaldehyde. The kidneys, lungs and attached trachea, and abdominal aorta were harvested within 30 min of death and fixed in 4% paraformaldehyde at 4 o C for at least 5 days. Then, the kidneys and aortas were immersed in 1 % alizarin red S (Sigma-Aldrich, St. Louis, MO) (pH 6.4) solution overnight and the lungs were placed in 0.2 % alizarin red S solution (pH 6.4) overnight. The stained tissues were transferred to saline, washed, and kept at 4 o C for at least 3 days. The saline was replaced at least 5 times during this washing step. For sectioning, the tissues stained with alizarin red were frozen in hexane at − 80 o C using a cooling apparatus (PSL-1800; Tokyo Rikakikai, Tokyo, Japan), and embedded in a cryo-embedding medium (Section-lab, Hiroshima, Japan). Ten-micrometer-thick sections were made using a Leica CM3050 S cryostat (Wetzlar, Germany) and dried on glasses at room temperature. Images were obtained using an Axio Scope A1 microscope (Carl Zeiss, Oberkochen, Germany) and AxioVision LE64 software (Carl Zeiss).

Statistical analysis
Statistical analysis was performed using the GraphPad Prism 8 statistical software (GraphPad Software, San Diego, CA). All data are presented as the mean ± standard deviation (SD). To compare multiple groups, one-way analysis of variance (ANOVA) with Tukey's post-hoc test or Dunnett's post-hoc test was performed. A P-value of < 0.05 was considered significant.

Down-regulation of serum PTH levels by a proresorptive dose of 1,25 (OH) 2 D 3
PTH, in association with vitamin D, plays a pivotal role in Ca and P homeostasis, and its synthesis is regulated by the Ca receptor and VDR signaling [3]. We previously showed the stimulatory effects of supra-physiological 1,25(OH) 2 D 3 administration on the osteoclast number, and serum carboxy-terminal cross-linked telopeptide of type 1 collagen (CTX-1, a bone resorption marker), Ca, and intact FGF23 levels, and body weight loss were blunted in both neutralizing anti-RANKL antibody-pretreated C57BL/6 wild-type mice and Osx-VDR-cKO mice [8]. The half-life of the anti-RANKL antibody was shown to be greater than 14 days [20], which is much longer than the duration of Experiment 1 in the previous [8] and present study (Fig. 1A). Thus, we examined effects of anti-RANKL antibody and ablation of VDR in osteoblastic cells on serum PTH levels ( Fig. 1A-D). 1,25(OH) 2 D 3 down-regulated serum PTH levels in control IgG-pretreated C57BL/6 mice (Fig. 1C) and Osx-VDR-wild-type (control) mice (Fig. 1D). Unlike changes in osteoclast number, serum CTX-1, FGF23, and Ca levels, and body weight [8], serum PTH levels were down-regulated by 1,25 (OH) 2 D 3 administration in anti-RANKL antibody-pretreated C57BL/6 mice and Osx-VDR-cKO mice in the same way as control IgG-pretreated C57BL/6 mice and Osx-VDR-wild-type (control) mice.

Soft tissue calcification induced by a proresorptive dose of 1,25 (OH) 2 D 3
Possible associations between excessive bone resorption and ectopic calcification have been proposed under many pathological conditions and in disease models, such as vitamin D-treated rodents [32][33][34], warfarin-treated rodents [32,33], CKD model rodents [30,34], osteoprotegerin-deficient mice [35], and patients with hypervitaminosis D [19], CKD [36,37], and postmenopausal osteoporosis [38]. Administration of a proresorptive dose of 1,25(OH) 2 D 3 upregulated serum Ca levels in all wild-type mice, while serum P levels tended to increase but were not always significantly elevated by this treatment [8]. High serum Ca x P products are known as biological predispositions to ectopic calcification of soft tissues and lead to high mortality in CKD patients [39,40]. Thus, we examined whether administration of a proresorptive dose of 1,25(OH) 2 D 3 induces soft tissue calcification, and if so, whether soft tissue calcification and concomitant elevation of the serum Ca x P product are prevented by treatment with anti-RANKL antibody or VDR ablation in osteoblastic cells. Treatment for 4 days with 1,25(OH) 2 D 3 led to marked alizarin red staining for calcification in the kidneys, lungs, and aorta ( Fig. 2A and B), consistent with previous reports using vitamin D [32,33]. We confirmed that administration of 1,25(OH) 2 D 3 elevated the serum Ca x P product ( Fig. 2A and B, right panels). Then, the effects of anti-RANKL antibody injection and ablation of VDR in osteoblastic cells on soft tissue calcification and the serum Ca x P product were tested in mice treated for 4 days with 1,25(OH) 2 D 3 . Like 1,25(OH) 2 D 3 -induced changes in the osteoclast number, serum CTX-1, Ca, FGF23 levels, and body weight loss in our previous study [8], 1,25(OH) 2 D 3 -induced calcification in the kidneys, lungs, and aorta and elevation of the serum Ca x P product were also prevented by a single injection of anti-RANKL antibody and inactivation of the VDR gene in osteoblastic cells ( Fig. 2A  and B). Microscopic examination of alizarin red-stained sections revealed calcification of the cortex of the kidneys, bronchioles, alveoli, and blood vessels in the lung, and the elastic lamellae in the media of the aorta in the 1,25(OH) 2 D 3 -treated mice. Such 1,25(OH) 2 D 3 -induced calcification was largely absent in the tissues from anti-RANKL antibody-pretreated mice and Osx-VDR-cKO mice ( Fig. 3A and B).
We also analyzed mRNA expression of a local calcification stimulator, tissue-nonspecific alkaline phosphatase (Alpl) [41], and calcification inhibitors such as matrix Gla protein (Mgp) [42] and ectonucleotide pyrophosphatase/phosphodiesterase 1 (Enpp1) [43,44] in kidneys ( Supplementary Fig. 1A-C). Our previous study revealed that VDR expression levels in kidneys in Osx-VDR-cKO mice were similar to those in control mice [5]. Upregulation of Mgp and downregulation of Alpl and Enpp1 expressions by administration of 1,25(OH) 2 D 3 were observed in control IgG-pretreated mice but not in anti-RANKL antibody-pretreated mice. Such changes were not observed in control mice or Osx-VDR-cKO mice in Experiment 2. These results suggest that the systemic Ca x P product but not local expression of Mgp, Alpl, or Enpp1 accounts for the ectopic calcification induced by 1,25(OH) 2 D 3 .

Discussion
In our previous study, 1,25(OH) 2 D 3 -induced upregulation in RANKL expression, osteoclastogenesis, serum CTX-1, Ca, P, and FGF23 levels, and body weight loss were significantly blunted in anti-RANKL antibody-pretreated mice and Osx-VDR-cKO mice, compared with control animals [8] (Fig. 4A). In the present study, we demonstrated that pretreatment with anti-RANKL antibody and VDR ablation in osteoblastic cells prevented 1,25(OH) 2 D 3 -induced soft tissue calcification and simultaneous increase of the serum Ca x P product. We also demonstrated that 1,25(OH) 2 D 3 administration down-regulated serum PTH levels in both test groups of mice to the same extent as in the respective control mice. PTH is known to paradoxically cause the net loss (=Ca and P release from bone) and net increase (=Ca and P deposition) of bone [13,14]. Both effects of PTH are dosage regimen-dependent and possibly physiological secretion pulsatility-dependent [13,14,45]. The regimen and pulsatility are considered to affect conformational states of the PTH 1 receptor [14,15]. Thus, PTH may have been involved in preventive actions of anti-RANKL antibody and VDR deficiency in osteoblastic cells against toxic actions of vitamin D. However, down-regulation of PTH by 1,25(OH) 2 D 3 administration occurred regardless of whether the proresorptive action was inhibited. These results suggest that the regulation of PTH is not involved in prevention of vitamin D toxicity, and that 1,25 (OH) 2 D 3 directly inhibits PTH secretion from the parathyroid glands.
Soft tissue calcification induced by 1,25(OH) 2 D 3 administration may be regulated by local environmental factors. Warfarin treatment has been shown to promote vascular calcification, possibly by increasing an inactive form of MGP, uncarboxylated MGP [32]. In our experiment, renal expression of Mgp was little affected by 1,25(OH) 2 D 3 administration. Administration of osteoprotegerin (a decoy receptor for RANKL) suppressed vascular calcification in warfarin-treated rats without lowering serum Ca or P [33], suggesting that bone resorption is involved in warfarin-induced vascular calcification. Enpp1 is the principal enzyme that produces inorganic pyrophosphate (PPi), a potent inhibitor of calcification [43,44]. Alpl is an established marker of osteoblast differentiation [41]. Alpl hydrolyzes PPi to generate two molecules of P [41,44]. Therefore, transdifferentiation of vascular cells to osteoblast-like cells at local sites has been considered to induce ectopic calcification there [30,34]. We showed that neither Enpp1 nor Alpl expression in kidneys was largely affected by 1,25(OH) 2 D 3 administration. Thus, local factors such as Mgp, Alp, and Enpp1 may not be directly involved in the ectopic calcification induced by 1,25(OH) 2 D 3 .
Soft tissue calcification is common among late-stage CKD patients and CKD model animals [17,18,46]. The mechanism of soft tissue calcification observed in CKD may differ from that in hypervitaminosis D. In CKD patients and animal models, low serum Ca and high P levels are observed due to reduced Ca reabsorption and P excretion [47,48]. An adenine-induced CKD mouse model exhibits negative Ca balance and low serum 1,25(OH) 2 D 3 levels [30]. These results suggest that steady Although systemic 1,25(OH) 2 D 3 levels were low in CKD model mice, local osteocytic production of 1,25(OH) 2 D 3 was markedly enhanced [30]. Osteocyte-specific Cyp27b1 [the enzyme synthesizes 1,25 (OH) 2 D 3 ] cKO mice exhibited more severe extraskeletal calcification than control mice [30]. This suggests that osteocyte-generated 1,25 (OH) 2 D 3 has a protective role in soft tissue calcification in the CKD model through 1,25(OH) 2 D 3 -VDR signaling in osteoblastic cells. In contrast, our study indicates that the 1,25(OH) 2 D 3 -VDR signaling in osteoblastic cells enhances soft tissue calcification in mice with hypervitaminosis D. These results suggest that the downstream of 1,25 (OH) 2 D 3 -VDR signaling in osteoblastic cells in CKD may be different from that in hypervitaminosis D. In order to clarify different mechanisms of soft tissue calcification, it is necessary to compare the 1,25 (OH) 2 D 3 -VDR transcriptome in osteoblastic cells in CKD and hypervitaminosis D.
The mechanism responsible for the association between bone resorption and soft tissue calcification in hypervitaminosis D remains unknown. One possible mechanism for the association is that soft tissue calcification is a passive biochemical process triggered by high serum Ca and P levels. High serum Ca and P levels could stimulate spontaneous hydroxyapatite formation in soft tissues, if an unidentified nucleating substance of hydroxyapatite crystals exist at the site. Another possibility is that supraphysiological 1,25(OH) 2 D 3 -induced bone resorption facilitates the release of an ectopic calcification-stimulating factor, 'X', that may be a VDR target gene product in osteoblastic cells (Fig. 4B). Such a factor may promote the nucleation and growth of hydroxyapatite crystals in soft tissues when serum levels of Ca x P products are extremely high. In an adenine-induced CKD model, 1,25(OH) 2 D 3 -regulated osteocytic sclerostin and BMP2 were shown to modulate soft tissue calcification [30]. The ability of bone-derived factors including sclerostin and BMP2 to act as regulators of ectopic calcification will be analyzed in the future. Taken together, our results suggest that suppression of bone resorption is an effective means to prevent various symptoms such as soft tissue calcification and body weight loss observed in hypervitaminosis D.

Conclusions
We demonstrate that VDR in osteoblastic cells is crucial for bone resorption, hypercalcemia, high serum Ca x P products, FGF23 synthesis, body weight loss, and soft tissue calcification, induced by 1,25 (OH) 2 D 3 . VDR in osteoblastic cells plays a central role in hypervitaminosis D, and may also play an important role in other Ca imbalance disorders. We also show that serum PTH is not directly involved in 1,25 (OH) 2 D 3 -induced soft tissue calcification. These results suggest that 1,25 (OH) 2 D 3 -induced bone resorption promotes the release of an ectopic calcification-stimulating factor. The Osx-VDR-cKO mouse may be a useful tool to search for calcification-stimulating factors. This study also suggests that the use of bone resorption inhibitors may be a treatment option not only for patients with hypervitaminosis D but also for those with other diseases having complications of hypercalcemia and vascular calcification.

Disclosure summary
The authors have nothing to disclose.

Funding
This study was supported in part by JSPS grants: 19KK0234 and Fig. 4. Summary of effects of a supraphysiological dose of 1,25(OH) 2 D 3 observed in our previous and present studies, and a hypothetical mechanism. (A) A summary of the results obtained in our studies. In Mori et al. [8], we demonstrated that a high dose of 1,25 (OH) 2 D 3 increased serum Ca levels, bone resorption, and serum FGF23 levels, and decreased body weight in all mice, and increased P levels in some mice. Here, we showed that 1,25(OH) 2 D 3 decreased serum PTH levels, increased serum Ca x P products, and induced soft tissue calcification. High levels of serum Ca and Ca x P products, excessive bone resorption, the body weight loss, and soft tissue calcification are known as symptoms of hypervitaminosis D. These toxic outcomes were prevented in anti-RANKL antibody-pretreated wild-type mice and more prominently in Osx-VDR-cKO mice. Only down-regulation of serum PTH levels by 1,25(OH) 2 D 3 was not attenuated at all in anti-RANKL antibody-pretreated mice or Osx-VDR-cKO mice. (B) A proposed hypothetical mechanism of supra-physiological 1,25(OH) 2 D 3 . VDR is preferentially expressed in osteoblasts and osteocytes (osteoblastic cells) in bone. 1,25(OH) 2 D 3 acts on VDR in osteoblastic cells to enhance RANKL expression and production of an unknown factor, 'X', that promotes soft tissue calcification. Then, osteoclasts are induced and bone resorption is enhanced. The factor 'X' is released from sites of bone resorption, travels in blood to settle in soft tissue structures, and promotes transdifferentiation to osteoblast-like cells. Greater Ca and P availability from bone resorption results in calcification there. 20H03872 (to YN); 20K21689 and 21H03125 (to NU).

Data availability
No data was used for the research described in the article.

Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jsbmb.2023.106351.