Technetium-99 Conjugated with Methylene Diphosphonate Ameliorates Glucocorticoid Induced Osteoporosis by Inhibiting Osteoclastogenesis

Technetium-99 conjugated with methylene diphosphonate (99Tc-MDP) is an effective anti-inflammatory drug in treating rheumatoid arthritis (RA) for over 15 years in China. However, as a special form of bisphosphonate, the antiosteoporotic effect of 99Tc-MDP is unclear. We systematically investigated the effects of 99Tc-MDP on cancellous and cortical bone, respectively, in glucocorticoid induced osteoporosis (GIO) animal models. Forty-eight Sprague-Dawley rats were randomly divided into six groups: blank, negative control, high dose, medium dose, low dose, and positive control groups. After dexamethasone was given to all groups except the blank group to induce osteoporosis, the rats in different groups were treated with saline, MDP, or different doses of 99Tc-MDP. After treatment, all rats were sacrificed, and their tibiae and femora were analyzed with microcomputed tomography (micro-CT), histology and biomechanics. Micro-CT analyses showed that (1) 99Tc-MDP reversed glucocorticoid induced bone microarchitecture destruction by increasing BV/TV, Tb.Th, and Tb.N and decreasing BS/BV, Tb.Sp, and TBPf; (2) effect of 99Tc-MDP increased as its dosage increased; and (3) 99Tc-MDP could improve cortical bone thickness while MDP failed to do so. Micro-CT spatial structure analysis and histology also yielded consistent results, indicating that 99Tc-MDP increased trabecular number and connectivity morphologically. Secondly, biomechanics revealed that 99Tc-MDP can enhance the extrinsic stiffness of bone by changing bone geometry. Finally, 99Tc-MDP could inhibit osteoclastogenesis in PBMCs in human. In conclusion, 99Tc-MDP exerted antiosteoporotic effect by improving both cancellous and cortical bone, as well as increasing extrinsic bone stiffness which might be attributed to the its inhibition of osteoclast differentiation. The antiosteoporotic effect of 99Tc-MDP may suggest a potential clinical application for patients with GIO.


Introduction
Glucocorticoid (GC), with its fast and prominent anti-inflammatory and immunosuppressive effects, has been widely used in the treatment of immune-mediated diseases including rheumatoid arthritis (RA) [1]. GC therapy, even in low dose, can lead to significantly decreased radiographic progression of rheumatoid arthritis [2]. However, it can cause significant adverse effects such as glucocorticoid induced osteoporosis (GIO), a most common form of secondary osteoporosis [3,4]. Therefore, early intervention and management of GIO is crucial for patients receiving GC therapy [5].
While different medications have been developed to address GIO, previous researches have proven that bisphosphonate is particularly effective in preventing and treating glucocorticoid induced bone loss [6][7][8]. A specific form of bisphosphonate, technetium-99 conjugated with methylene diphosphonate ( 99 Tc-MDP), is the radioactive-safe decay product of 99 mTc-MDP which has no radioactivity and is harmless to human body in long term clinical treatment. It was patented in 1995 (patent No. ZL94113006.1) and approved by the State Food and Drug Administration (SFDA) of China in 2002 (Approval No. H20000218) to treat RA due to its antiinflammation efficacy.

BioMed Research International
Study in collagen induced arthritis (CIA) rats has shown that 99 Tc-MDP is effective in suppressing the production TNF-and IL-1 [9]. A very recent study has also shown the serum levels of TNF-and IL-6 are reduced significantly while the level of anti-inflammatory cytokine TGF-was increased in patients with RA who received 99 Tc-MDP therapy [10]. Interestingly, in their study, they firstly found the antiinflammation effects of 99 Tc-MDP might be exerted via upregulating the frequency of CD4+CD25+Foxp3+ Tregs and T cell in peripheral blood. Furthermore, a previous study demonstrated that 99 Tc-MDP decreased the rheumatoid factor concentration in RA patients which indicated 99 Tc-MDP could also control the immune activities of RA [11]. 99 Tc-MDP has been used in Chinese RA patients for over 15 years based on its anti-inflammation effects. And till now most studies have focused on this efficacy in RA. However, as a bisphosphonate, less attention has been paid to the antiosteoporotic effect [9,[11][12][13]. The objective of our present study is to systematically investigate the effect of 99 Tc-MDP on cancellous and cortical bone, respectively, in vivo by applying microcomputed tomography (micro-CT), histomorphology, and biomechanical test, as well as it on the osteoclast formation in vitro.

Animal Model.
Forty-eight three-month-old female Sprague-Dawley rats, weighing 256.96 ± 18.18g, were obtained from Beijing Vital River Laboratory Animal Technology Co. Ltd. The rats were kept in the same room of the laboratory animal center of Peking University Health Science Centre, fed on the same diet daily, and weighed weekly. The forty-eight rats were randomly and evenly divided into six groups: blank group, negative control group (saline), high dose group ( 99 Tc-MDP 10 mg/kg), medium dose group ( 99 Tc-MDP 5 mg/kg), low dose group ( 99 Tc-MDP 2.5 mg/kg), and positive control group (MDP 5 mg/kg).
During the induction period, physiological saline was given to the blank group and dexamethasone to the other five groups, 5 mg/kg, i.m, twice a week, for 16 weeks. In the treatment period, the blank and the negative control groups were given 5 mg/kg saline; 10 mg/kg, 5 mg/kg, and 2.5 mg/kg 99 Tc-MDP were given to the high dose, medium dose, and low dose groups, respectively; and the positive control group was given 5 mg/kg MDP, all intravenously, once a week, for 14 weeks. Dexamethasone was given in the same way mentioned above during the treatment period to maintain osteoporotic effect, except that the frequency was reduced to once a week.
Sterile injection of 99 Tc-MDP was prepared before use by mixing two preparations. Preparation A was transparent solution containing 0.05 g 99Tc. Preparation B was lyophilized powder containing 5 mg MDP and 0.5 mg stannous chloride. Both preparations were sealed in 5 mL glass vials and stored in darkness at 2-8 ∘ C.
The rats were sacrificed after treatment. Their femora and tibiae were collected for micro-CT analysis, histology, and biomechanical test as described in the following sections.
All animal treatments and procedures were approved by the ethical committee of Peking University Health Science Center.

Microcomputed Tomography
2.2.1. Microstructural Quantitative Analysis. The right tibiae were thawed in room temperature; the proximal part was scanned by micro-CT (Siemens Inveon CT/PET, Germany) in high resolution set by Inveon Acquisition Workplace. The voltage was 60 kV and the current was 400 A; the total rotation was 360 degrees at a rotation step of 2 degrees (integration time was 12 minutes). The effective pixel size was 10.34 m. The acquired scan data sets were reconstructed at a voxel size of 5.17 × 5.17 × 10.34 m 3 .
A 2 mm thick region, which was 1 mm distal to the proximal tibial growth plate, was defined as a volume of interest (VOI) for quantitative analysis with Inveon Research Workplace. Six parameters of cancellous bone were also calculated by the workstation: bone volume/total volume (BV/TV), bone surface area/bone volume (BS/BV), trabecular thickness (Tb.Th), trabecular number (Tb.N), trabecular spacing (Tb.Sp), and trabecular bone pattern factor (TBPf). One parameter of cortical bone was calculated: cortical thickness (Ct.Th) by measurements at 8 different sites.

Spatial Structure of the Cancellous Bone.
A 0.4 mm thick region, which was 1.8 mm distal to the proximal tibial growth plate, was defined as a VOI for 3D construction to show the spatial structure of cancellous bone.

Histology.
The left tibiae were fixed with 4% paraformaldehyde, embedded in paraffin to make 4-m-thick sections, then stained with hematoxylin and eosin (H & E), and observed under light microscope.

Biomechanical Tests
2.4.1. Extrinsic Biomechanical Properties. Right femora were thawed before three-point bending test and submerged in a saline bath maintained at 37 ∘ C. Load was applied midway between a 15-mm loading span at a speed of 6 mm/s using mechanical testing machine (MTS Systems Corp., Eden Prairie, MN, USA). Fu (ultimate force), Du (ultimate displacement), S (stiffness), and U (energy absorbed) were determined from the load-displacement curve recorded by software [14].

Bone Geometry.
Bone geometry measurements of the midshaft of the fractured femora were performed using digital calipers (accuracy: 0.005 mm) after three-point bending test, including width of the bone in the mediolateral direction, (a) width of the bone in the anteroposterior direction (b) and cortical thickness (Ct.Th). Average cortical thickness was calculated from thickness measurements made in each of the four directions of the femoral cross-section.
The value for the cross-sectional moment of inertia (CSMI) at the midshaft of the femur was calculated using the elliptical cross-section model [15]: (1)

Intrinsic Biomechanical Properties.
Intrinsic biomechanical properties were calculated from the following formulas: where u is ultimate stress, u is ultimate strain, E is elastic modulus, is modulus of toughness, L is the loading span (15 mm), and I is the CSMI [16].

Osteoclast Differentiation Assay.
A total of 7 rheumatoid arthritis (RA) patient peripheral blood mononuclear cells (PBMCs) were isolated from fresh heparinized venous blood samples using Ficoll density-gradient centrifugation and then were plated in a-MEM medium (Life Technologies, Grand Island, NY, USA) supplemented with 10% FBS (Life Technologies) at 5× 10 4 cells per 200 l per well in 96-well plates. After adhering for 4 h in the incubator, nonadherent cells were removed to get the adherent osteoclast precursors (>90% CD14+) under the stimulation of 30 ng/ml recombinant human macrophage colony stimulating factor (rhM-CSF, Peprotech GmbH, Rocky Hill, CT, USA)) and 50 ng/ml recombinant human receptor activator of nuclear factorkappa B ligand (rhRANKL, R&D systems, Minneapolis, MN, USA) with or without 100 ng/ml rhTyro3 Fc chimera. Medium was changed every 3 days. On day 17, the cells were detected by tartrate-resistant acid phosphatase (TRAP) staining with the Leukocyte Acid Phosphatase kit (Sigma-Aldrich, St. Louis, MO, USA) and TRAP positive multinucleated cells (three or more nuclei) were counted under an inverted fluorescence microscope (Olympus IX71-141, Tokyo, Japan).

Statistical Analysis.
All statistical analyses were performed using SPSS 17.0. Data were tested for normality and homogeneity of variances. Results were expressed as ± s and tested for significance using independent or paired t-test. Simple linear regression and Pearson's correlation coefficient were calculated to test for linear correlations. P value small than 0.05 was considered statistically significant.

99 Tc-MDP Improved BV/TV, Tb.N, BS/BV, Tb.Sp, and TBPf
Significantly in Cancellous Bone, as well as Ct.Th in Cortical Bone in the Rats with GIO. Microstructural quantitative analysis of cancellous bone by microcomputed tomography showed that, (1) compared with the blank group, rats in the negative control group showed significant decreases in BV/ TV and Tb.N as well as significant increases in BS/BV, Tb.Sp, and TBPf (p < 0.05); (2) compared with the negative control group, rats in the low dose group showed significant increases in BV/TV and Tb.N and significant decreases in Tb.Sp and TBPf (p < 0.05); (3) for the medium dose and high dose groups, the parameters of Tb.Th and BS/BV showed significantly improvement as the dosing increased; and (4) for the positive control group, improvement of all six parameters reached statistical significance (Table 1 and Figure 1).
Quantitative analysis of cortical bone revealed that (1) Ct.Th decreased in the negative control group in contrast with the blank group (p < 0.05); (2) after treatment, Ct.Th in three 99 Tc-MDP treatment groups, regardless of dose differences, all increased significantly, comparing with the negative control group (p < 0.05); (3) no significant improvement of Ct.Th was observed in the positive control group, when compared with the negative control group (Table 1 and Figure 1).

99 Tc-MDP Ameliorated the Disorder in Structure of Cancellous Induced by Glucocorticoid.
Structures of cancellous from all the groups were evaluated by 3D structure in the present study. The results showed that, compared with the blank group, trabeculae in the negative control group became thinner and sparse, and the microstructure was compromised. The density, integrity, and connectivity of trabeculae in high dose group, medium dose group, low dose group, and positive control group all improved substantially compared with the negative control group (Figure 2, left panel).
H&E staining was also performed in our present study. Our finding showed that trabeculae in the negative control group became sparse and disordered compared with the blank group; by contrast, trabeculae in high dose group, medium dose group, low dose group, and positive control group all became denser and regular, which was consistent with the 3D structure mentioned above ( Figure 2, right panel).

The Efficacy of 99 Tc-MDP on Stiffness, Cortical Thickness, and Cross-Sectional Moment of Inertia.
Compared with the blank group, stiffness (S), cortical thickness (Ct.Th), and cross-sectional moment of inertia (CSMI) all decreased in the negative control group. However, only the decreases in S and Ct.Th were statistically significant. In addition, all the three 99 Tc-MDP treatment groups, when compared with the negative control group, showed increases in S and Ct.Th (P < 0.05), but only the high dose group showed significant increase in CSMI. S and Ct.Th increased in the positive control group, but there was no statistical significance. The positive control group showed significant increase in CSMI (p < 0.05) (Table 2 and Figure 3).
However, ultimate force (Fu), energy absorbed (U), elastic modulus (E), and modulus of toughness ( T) did not vary significantly between the groups (Table 2).

99 Tc-MDP Inhibited Osteoclastogenesis in Human PBMCs
In Vitro. To further investigate the underlying mechanism  Data are presented as mean ± standard deviation. Extrinsic parameters: Fu (ultimate force), Du (ultimate displacement), S (stiffness), and U (energy absorbed); bone geometry: b (anterior-posterior width), a (media-lateral width), Ct.Th (cortical thickness), and CSMI (cross-sectional moment of inertia); intrinsic parameters: u (ultimate stress), u (ultimate strain), E (elastic modulus), and T (modulus of toughness). All the other five groups are compared with the negative control group (saline group), respectively. * p < 0.05.
of 99 Tc-MDP on GIO. We evaluated the osteoclast inhibition effect of 99 Tc-MDP in human PBMCs in vitro by TRAP staining. Seven RA patients were enrolled in our present study. PBMCs were isolated and stimulated with RANKL and M-CSF to promote macrophage differentiated to osteoclasts. All groups except the blank group were treated with 2 g/ml or 10 g/ml 99 Tc-MDP, respectively. 99 Tc-MDP with a concentration of 2 g/ml showed no significant inhibition, but with 10 g/ml exhibited complete inhibition on osteoclastogenesis (Figures 4(a) and 4(b)).

Discussion
In the present study, as we known, we are the first to show that 99 Tc-MDP ameliorates glucocorticoid induced osteoporosis in vivo. We demonstrate the anti-GIO effects of 99 Tc-MDP by microcomputed tomography, bone histomorphology, and biomechanics in vivo and antiosteoporosis by TRAP staining in vitro. Our findings indicate that 99 Tc-MDP effectively ameliorates glucocorticoid induced osteoporosis by inhibiting osteoclastogenesis.
Micro-CT, as our major investigation tool to quantify bone microstructure, was able to deliver detailed and accurate anatomical information and measurements, due to its ability to achieve high spatial resolution [17,18]. It was first introduced by Feldkamp, L.A., to directly examine three-dimensional bone structure in vitro; structural indices commonly determined from two-dimensional histological sections can be obtained nondestructively from a large number of slices in each of three orthogonal directions with little or no preparation of the sample [18][19][20]. Since then, this technology has become a powerful tool for assessing the 3D architecture of trabecular bone. Measurements from micro-CT were excellently correlated with those measured from conventional histomorphometry and therefore were very reliable [18,[21][22][23][24][25].
Our study had three important findings. First, low dose of 99 Tc-MDP (2.5 mg/kg/week) was able to significantly reverse the microstructural change of cancellous and cortical bone caused by long term glucocorticoid use. Our research is the first one to confirm that 99 Tc-MDP may not only improve cancellous bone, demonstrated by increased BV/TV and Tb.N and decreased Tb.Sp and TBPf, but also thicken cortical thickness impaired by glucocorticoid. Second, besides quantitative evidence described above, morphological evidence derived from both micro-CT 3D structure and histology of cancellous bone consistently showed that 99 Tc-MDP treatment yielded denser, thickened trabeculae, and more integrated cancellous structure compared with the negative control group. Third, 99 Tc-MDP demonstrated capability to change bone geometry such as increasing cortical thickness and CSMI, therefore to enhance bone extrinsic stiffness in biomechanical test. There was no significant intrinsic stiffness change after 99 Tc-MDP treatment.
By comparing effects of 99 Tc-MDP and MDP, we found that MDP also showed excellent ability to improve cancellous bone. However, it has no effect on cortical bone thickness (as shown in both micro-CT measurements and digital caliper The differences between every two groups were evaluated by independent test. A p < 0.05 was considered to be statistically significant. measurements) and bone extrinsic stiffness impaired by glucocorticoid.
As for the underlying antiosteoporotic mechanism, the P-C-P structure of MDP enables 99 Tc-MDP to localize in bone and joint after injection, then inhibits activity of osteoclasts and bone absorption and promotes osteoblastogenesis and bone formation, which may be one of the most important antiosteoporotic mechanism of 99 Tc-MDP. In addition, microelement 99 Tc has a long half-life in human body and therefore affords 99 Tc-MDP to exert its therapeutic effects in a long term; this characteristic is distinct from conventional drug relying on plasma concentration to play a role and is probably the reason why 99 Tc-MDP appears to be more effective than MDP on improving cortical bone.
To better understand the antiosteoporotic mechanism of 99 Tc-MDP, we performed the in vitro study in PBMCs from RA patients. We found 99 Tc-MDP could inhibit M-SCF and RANKL induced osteoclastogenesis which was consistent with a recent study of effect of 99 Tc-MDP on ovariectomyinduced osteoporotic in mice [13]. In their study, they also found the 99 Tc-MDP could upregulate CD4+CD25+Foxp3+ Tregs and downregulate Th17 cells in spleen in ovariectomyinduced osteoporotic mice. The upregulation of CD4+ CD25+Foxp3+ Tregs by 99 Tc-MDP was also found by Wu and her colleagues in peripheral blood from RA patients [12]. Were CD4+CD25+Foxp3+ Tregs involved in the development of osteoporosis? Tregs were hypothesized to exert the inhibitory effects on osteoclast formation in several studies [26][27][28]. It suggested that the inhibition of osteoclast formation by 99 Tc-MDP might be a result of the upregulation of Tregs.
In conclusion, the data presented here provide new insight into the underlying effects of 99 Tc-MDP in improving both cancellous and cortical bone, and increasing extrinsic bone stiffness might be via inhibiting osteoclastogenesis. However, the current data are insufficient to determine whether the antiosteoporotic effect of 99 Tc-MDP, in patients with GIO, is as effective as it in vivo and in vitro. Hence, ongoing studies have been designed to assure the antiosteoporotic effects of 99 Tc-MDP in patients with GIO.