Osteoporos Int DOI 10.1007/s00198-016-3781-6

ORIGINAL ARTICLE

Autoimmune arthritis deteriorates bone quantity and quality of periarticular bone in a mouse model of rheumatoid arthritis T. Shimizu 1 & M. Takahata 1 & H. Kimura-Suda 2 & Y. Kameda 1 & K. Endo 1,3 & H. Hamano 1 & S. Hiratsuka 1 & M. Ota 1 & D. Sato 1 & T. Ito 2 & M. Todoh 3 & S. Tadano 3 & N. Iwasaki 1

Received: 9 June 2016 / Accepted: 15 September 2016 # International Osteoporosis Foundation and National Osteoporosis Foundation 2016

Abstract Summary This study showed that autoimmune arthritis induces especially severe osteoporosis in the periarticular region adjacent to inflamed joints, suggesting that arthritis increases the fragility fracture risk near inflamed joints, which is frequently observed in patients with RA. Introduction Periarticular osteoporosis near inflamed joints is a hallmark of early rheumatoid arthritis (RA). Here we show that rheumatic inflammation deteriorates the bone quality and bone quantity of periarticular bone, thereby decreasing bone strength and toughness in a mouse model of RA. Methods Female BALB/c mice and SKG mice, a mutant mouse model of autoimmune arthritis on the BALB/c background, were used. At 12 weeks of age, BALB/c mice underwent either Sham surgery or bilateral ovariectomy (OVX), and SKG mice underwent intraperitoneal injection of mannan to induce arthritis. Eight weeks later, the mice were killed and the femurs and tibias were subjected to micro-computed tomography, Fourier transform infrared (FTIR) spectroscopic imaging, X-ray diffraction, histology, and mechanical testing. Results SKG mice developed significant trabecular bone loss in both the distal metaphysis of the femur and the lumbar vertebral body, but the extent of the bone loss was more severe in the distal metaphysis. Neither SKG nor OVX mice

* M. Takahata [email protected]

1

Department of Orthopedic Surgery, Hokkaido University Graduate School of Medicine, Kita-15 Nishi-7, Kita-ku, Sapporo 060-8638, Japan

2

Chitose Institute of Science and Technology, Chitose, Japan

3

Division of Human Mechanical Systems and Design, Faculty of Engineering, Hokkaido University, Sapporo, Japan

exhibited changes in the geometry and matrix properties of the diaphysis of the femur, whereas SKG mice, but not OVX mice, did exhibit changes in these properties in the distal metaphysis of the femur. Bone strength and fracture toughness of the distal metaphysis of the tibia adjacent to the inflamed ankle joint were significantly decreased in SKG mice. Conclusions Autoimmune arthritis induces periarticular osteoporosis, characterized by deterioration of cortical bone geometry and quality as well as by trabecular bone loss, leading to severe bone fragility in periarticular bone adjacent to inflamed joints.

Keywords Bone fragility . Bone quality . Cortical bone . Rheumatoid arthritis (RA)

Abbreviations RA Rheumatoid arthritis RANKL Receptor activator of NF-kB ligand SH2 Src homology 2 ZAP-70 ζ-associated protein of 70 kDa CTX C-terminal telopeptides of type I collagen BV/TV Trabecular bone volume fraction Tb.N Trabecular number Tb.Th Trabecular thickness SMI Structural model index Ct.Th Cortical thickness FTIR Fourier transform infrared PMMA Poly methyl methacrylate MCT Mercury–cadmium–telluride MTMR Mineral to matrix ratio CTPR Carbonate to phosphate ratio HAp Hydroxyapatite SD Standard deviation

Osteoporos Int

Introduction Rheumatoid arthritis (RA) is a systemic inflammatory disorder characterized by symmetrical polyarthritis that affects ∼1 % of the world population [1]. The osteoclast pathway is activated by abnormal immune conditions in association with chronic synovial inflammation, resulting in focal marginal articular erosion, subchondral bone loss, periarticular osteoporosis [2], and systemic osteoporosis [3, 4]. Among these bone lesions, periarticular osteoporosis is a hallmark of early RA [5, 6]. Juxta-articular bone loss precedes marginal joint bone erosion, and therefore, periarticular osteoporosis is used as a radiologic predictor for the subsequent development of marginal joint erosion [7, 8]. Periarticular osteoporosis in RA is caused by increased bone resorption resulting from the accumulation of inflammatory cells in the bone marrow of the juxta-articular region [9]. Inflammatory cells, including lymphocytes and macrophages, are a likely source of receptor activator of NF-kB ligand (RANKL) as well as proinflammatory cytokines, which promote osteoclast-mediated bone resorption [10–12]. Additionally, reduced mechanical loading due to the immobilization of painful arthritic joints facilitates the development of periarticular osteoporosis [13]. Rapid bone loss resulting from inflammation and immobilization is mainly responsible for decreased periarticular bone strength; however, abnormal bone metabolism induced by inflammation and immobilization may also lead to the deterioration of periarticular bone quality. Qualitative changes of the bone in RA are of great interest because patients with RA experience fractures more frequently than expected based on predictions of bone mineral density [14]. Osteoporosis in patients with RA is thought to cause particularly severe bone fragility due to bone quality abnormalities in addition to bone loss. To our knowledge, however, few studies have examined the qualitative bone changes in periarticular bone in RA. In the present study, we assessed changes in the bone quantity and quality of periarticular bone in close proximity to inflamed joints using an SKG mouse model of autoimmune arthritis [15] and compared them to the changes observed in an ovariectomized (OVX) mouse model of postmenopausal osteoporosis.

protein of 70 kDa (ZAP-70), a key signal transduction molecule in T cells [15]. At 12 weeks of age, SKG mice were injected intraperitoneally with 20 mg mannan (SigmaAldrich, St. Louis, MO, USA) to induce arthritis. Female BALB/c mice were OVX or Sham-operated (Sham) at 12 weeks of age. Arthritis was scored twice weekly based on the SKG scale. At 8 weeks after surgery or arthritis induction, the mice were killed by cervical dislocation, and blood samples and lower limbs were obtained for analysis. The Ethics Review Committee for Animal Experimentation of Hokkaido University approved the experimental protocol of this study. Bone marker analysis A serum assay for C-terminal telopeptides of type I collagen (CTX-I) was performed on the blood collected just before the mice were killed. Serum was separated from the blood obtained from fasted animals via centrifugation (30 min, 3000 rpm) at 4 °C and stored in single-use aliquots at −80 °C until analysis. An enzyme-linked immunosorbent assay kit for serum CTX-I was purchased from Immunodiagnostic System Inc. (Scottsdale, AZ, USA). Micro-computed tomography analysis

Methods

Right femurs and tibiae, as well as fifth lumbar vertebral bodies were scanned individually by micro-CT (R_mCT2; Rigaku, Tokyo, Japan) at a 10-μm isotropic resolution, and microstructural indices were measured in accordance with the guidelines described in Bouxsein et al. [16]. A 1500-μm area of interest of 150 slices encompassing the region of the distal metaphysis of the femur, starting from 500 μm proximal to the growth plate, was used to assess periarticular bone morphology. For the fifth lumbar vertebral body, an area from the upper growth plate to the lower growth plate was used to assess trabecular bone morphology. Trabecular bone parameters, including the trabecular bone volume fraction (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), structural model index (SMI), and degree of anisotropy (DA), were determined using an automatic algorithm of the TRI/3D-BON (Ratoc System Engineering Co., Tokyo, Japan). For cortical bone analysis, cortical thickness (Ct.Th) and polar moment of inertia of the same region were determined. Additionally, we measured CtTh of the breakdown point of the distal tibia for biomechanical testing.

Animals and arthritis induction

Fourier-transform infrared spectroscopic imaging

Female SKG mice were purchased from Japan Clea, and BALB/c mice were purchased from Jackson Laboratory. The mice were maintained under specific pathogen-free conditions. The SKG mouse model of autoimmune arthritis has a BALB/c background with a recessive point mutation of the gene encoding an Src homology 2 (SH2) domain of ζ-associated

The left femora were fixed in 70 % ethanol and subjected to undecalcified tissue processing. The specimens were embedded in polymethyl methacrylate (PMMA; Wako Chemicals, Kanagawa, Japan), sectioned at 3 μm in the coronal plane, and mounted on BaF2. Spectra were acquired with a Spectrum Spotlight 400 Imaging System (Perkin Elmer, Waltham, MA,

Osteoporos Int

USA), which comprised an FTIR spectrometer (Spectrum 400) and an infrared (IR) microscope with a 16 × 1 element mercury– cadmium–telluride array detector. Infrared images were collected in the transmission mode at a spectral resolution of 4 cm−1 in the frequency region from 4000 to 680 cm−1 with an IR detector pixel size of 25 μm × 25 μm. The background spectra were collected through the BaF2 window. FTIR spectra were extracted from the FTIR images to determine bone quality and then the FTIR spectrum after subtraction of a linear baseline and PMMA spectrum was used to characterize the bone quality: mineral/matrix ratio (MTMR; PO43 − /amide I), which describes the degree of calcium phosphate mineralization; carbonate/phosphate ratio (CO32−/PO43−), which describes the amount of carbonate substitution in the apatite crystal lattice; and crystallinity of hydroxyapatite (1030 cm−1/1020 cm−1) [17]. To characterize bone quality, the following band areas were used: amide I (1710–1591 cm−1), PO43− (1186–906 cm−1), and CO32− (894–853 cm−1). FTIR images of MTMR were displayed using Spectrum IMAGE (PerkinElmer, MA, USA) after subtracting the PMMA area.

All sections were counterstained with methyl green and observed under light microscopy. Biomechanical testing A three-point bending breakdown test was conducted at the distal tibia using a universal mechanical load-testing machine (Model 3365, Instron Corp., Norwood, MA, USA). The distal tibia was placed on the two lower support bars (the support bars were 4 mm apart), and the loading bar was positioned 7 mm from the distal end of the tibia. The initial load was 1 N and then a bending load was continuously applied at a crosshead speed of 0.5 mm/min until the bone broke. The maximum stress and toughness were determined from the load–deformation curve, and the stiffness and elastic modulus were determined from the load–deformation curve between 5 and 10 N. To calculate the elastic modulus and bending strength from the stress–strain curves, the bending moment was calculated by approximating an elliptical cylinder, which was based on a mean roundness of the cross sections of 0.85 ± 0.13 (calculated from 5 to 9 mm from the distal end of the tibia).

Infrared dichroism image analysis X-ray diffraction test Two different FTIR images of each sample were collected in the transmission mode at a spectral resolution of 4 cm−1 in the frequency region from 4000 to 680 cm−1 with an IR detector (pixel size 6.25 μm × 6.25 μm) using a wire grid polarizer. The IR dichroism image was defined as the ratio of the particular band areas when the specimen was determined using a parallel (0°) and perpendicular (90°) polarized IR beam and calculated using the following equation: D = (A∥ − A⊥)/(A∥ + A⊥), where A∥ and A⊥ were the particular band areas at 0° and 90°, respectively. The IR dichroism image of the collagen fibers was assessed by calculating the amide I band areas (1712–1601 cm−1) at 0° and 90° and displayed using Spectrum IMAGE (the area oriented in the direction of the bone axis is shown in red).

After mechanical testing, the distal tibia was irradiated for 10 min using an X-ray diffractometer (UltimaIV, Rigaku,

Histology For histological analysis, undecalcified PMMA sections of left femurs were stained with toluidine blue O (Merck, Darmstadt, Germany) and observed under light microscopy. We also performed immunohistochemistry for dentin matrix protein 1 (DMP-1) to identify changes in the osteocyte condition after exposure to inflammation. For immunohistochemistry for DMP-1, after pre-incubation with 1 % BSA–PBS for 30 min at room temperature, the sections were incubated with rabbit antibody against DMP-1 (Takara Bio Inc., Otsu, Japan) at a 1:500 dilution overnight at 4 °C. Sections treated with DMP1 antibody were then incubated with horseradish peroxidaseconjugated goat anti-rabbit IgG (DakoCytomation, Glostrup, Denmark) [18]. For visualization of all immunoreactions, diaminobenzidine tetrahydrochloride was used as a chromogen.

Fig. 1 Comparisons of the bone phenotype between BALB/c and SKG/ JCL mice at the start of the study. a Comparisons of bone volume and microstructural indices of trabecular and cortical bone of the femur between BALB/c and SKG/JCL mice are shown. BV/TV bone volume/ total volume, TbTh trabecular thickness, TbN trabecular number, CtTh cortical thickness. b Representative FTIR images of the mineral/matrix ratio (PO43−/amide I) of the distal femur in BALB/c mice and SKG/JCL mice at 12 weeks old

Osteoporos Int

Japan) with Mo-Kα characteristic X-rays and X-ray imaging plate (BAS-SR 127 × 127 mm2, Fujifilm, Japan). The Debye ring of the (002) planes of hydroxyapatite crystals was obtained in the X-ray imaging plate, and was calculated, which reflects the degree of c-axis orientation of hydroxyapatite crystals in the irradiated plane of the specimen [19]. Statistical analysis Comparisons among groups were performed using a one-way analysis of variance and Newman–Keuls tests. A significance level of P less than 0.05 was used for all comparisons. Data are represented as mean ± standard deviation (SD). All statistical analyses were performed using GraphPad Prism version 5 (GraphPad Software, San Diego, CA, USA).

Fig. 2 Enhanced bone resorption due to arthritis, leading to generalized bone loss. a Arthritis score in mice after surgery or arthritis induction are shown. Data are means ± SD from Sham BALB/c (open circle), OVX BALB/c (open triangle), and SKG/JCL (filled circle) mice. b Serum concentrations of CTX-I were significantly higher in the SKG group than in the Sham and OVX groups. c Micro-CT images of the fifth lumbar vertebral body of Sham and SKG mice at 8 weeks after surgery or arthritis induction. Comparisons of bone volume and microstructural indices of trabecular bone of the fifth lumber vertebral body of Sham and SKG mice are shown. vBMD volumetric bone mineral density of mineralized tissue, BV/ TV bone volume/total volume, TbTh trabecular thickness, TbN trabecular number, SMI structure model index, DA degree of anisotropy. Data represent means ± SD (*p < 0.05 vs Sham, †p < 0.05 vs OVX)

Results Bone phenotype of BALB/c and SKG/JCL mice at 12 weeks of age We first compared bone phenotype between BALB/c and SKG/JCL mice before OVX or arthritis induction. Bone volume and structure of the trabecular bone and cortical bone in the distal metaphysis of the femur assessed at 12 weeks of age using micro-CT, but BV/TV, Tb.Th, Tb.N, and Ct.Th, were not significantly different between the BALB/c and SKG/JCL mice (Fig. 1a). FTIR imaging was performed to measure the biochemical composition of the distal femur, but the mineral content (PO43−/amide I) did not differ significantly between the BALB/c and SKG/JCL mice (Fig. 1b).

Osteoporos Int

Arthritis score and systemic bone resorption A single intraperitoneal injection of 20 mg mannan was used to induce arthritis in SKG mice at 12 weeks of age. Arthritis developed in both the small and large joints at 1 week after injection, and arthritis activity peaked 6 weeks after injection, similar to previous reports [20]. The arthritis score was significantly higher in the SKG group than in the Sham and OVX groups from week 2 after arthritis induction to the end of the study, as previously described [21] (Fig. 2a). To evaluate the effect of arthritis or OVX on systemic bone resorptive activity, we measured serum CTX-I concentrations. The serum CTX-I concentration was significantly higher in the SKG group than in the Sham and OVX groups (Fig. 2b). Although not statistically significant, the serum CTX-I concentration tended to be higher in the OVX group than in the Sham group. Micro-CT evaluation of the fifth lumbar vertebral body was performed to assess generalized bone loss. The CT images showed that both OVX and SKG groups developed trabecular bone loss in the fifth lumbar vertebral body compared to the Sham group. The SKG group had a significant decrease in volumetric bone mineral density, BV/TV, Tb.Th, and Tb.N and an increase in SMI compared to the Sham and OVX Fig. 3 Arthritis-induced bone loss and structural deterioration of cortical bone as well as trabecular bone. a Representative micro-CT images of distal femurs in Sham, OVX, and SKG groups at 8 weeks after surgery or arthritis induction. b Comparisons of bone volume and microstructural indices of trabecular and cortical bone of the femur in Sham, OVX, and SKG mice are shown. BV/TV bone volume/total volume, TbTh trabecular thickness, TbN trabecular number, SMI structure model index, CtTh cortical thickness, J polar moment of inertia. Values shown are means ± SD (*p < 0.05 vs Sham, †p < 0.05 vs OVX)

groups (Fig. 2c). The OVX group also exhibited a decrease in volumetric bone mineral density, BV/TV, Tb.Th, and Tb.N and an increase in SMI, but the differences were not statistically significant.

Effects of arthritis on cortical and trabecular bone volume and structure of periarticular bone Bone volume and structure of the trabecular bone and cortical bone of the distal metaphysis of the femur were assessed at 8 weeks after arthritis induction or OVX by micro-CT. In general, SKG mice and OVX mice exhibited significant bone loss compared to Sham mice and the extent of bone loss was more severe in SKG mice than in OVX mice (Fig. 3a). Quantitative analysis of trabecular bone revealed a statistically significant decrease in BV/ TV, Tb.Th, and Tb.N and an increase in SMI, SKG, and OVX mice compared with Sham mice. Arthritis had a statistically significant effect on Ct.Th and polar moment of inertia in cortical bone, while OVX had no significant effect on cortical bone parameters (Fig. 3b). These findings indicate that arthritis has an especially strong deteriorating effect on the periarticular bone adjacent to inflamed joints.

Osteoporos Int Fig. 4 Arthritis-induced deterioration of the bone matrix quality of cortical bone near the arthritic joint. a Representative FTIR images of mineral to matrix ratio (PO43−/amide I) of the distal femur in the Sham, OVX, and SKG groups at 8 weeks after surgery or arthritis induction. b Comparisons of the quantitative parameters of the bone material properties calculated from the FTIR spectrum at the distal metaphysis of the femur. c Representative IR dichroism images, depicting collagen orientation, of the cortical bone of the distal femur in the Sham, OVX, and SKG groups at 8 weeks after surgery or arthritis induction. d A comparison of the area ratio per unit, i.e., the area in which collagen fibers are well oriented in the direction of the bone axis (range, 1.5–3.0) at the distal metaphysis and diaphysis of the femur. e Degree of hydroxyapatite (HAp) orientation in the cortical bone of distal tibias from the Sham, OVX, and SKG groups at 8 weeks after surgery or arthritis induction. f Representative histological images of the cortical bone of the distal femur in SKG mice at 8 weeks after arthritis induction. Left panels: toluidine blue; right panels: DMP-1; upper panels: metaphysis; lower panels: diaphysis. Black arrows show osteocytes. Values shown are means ± SD (*p < 0.05 vs Sham, †p < 0.05 vs OVX)

Osteoporos Int

Effects of arthritis on FTIR imaging-based bone qualitative parameters of periarticular bone FTIR imaging was performed to measure the biochemical composition of the distal femur. FTIR images of the mineral content (PO43−/amide I) showed that neither arthritis nor ovariectomy significantly affected the mineral content of the cortical bone at the diaphysis of the femur, but the mineral content of cortical bone at the distal metaphysis of the femur was decreased in SKG mice (Fig. 4a). The MTMR value at the diaphyseal region of the femur was similar among the three groups, but the MTMR value at the metaphyseal region was significantly lower in SKG mice than in Sham mice, indicating a reduction in the degree of mineralization at the periarticular bone adjacent to the inflamed joints (Fig. 4b). The carbonate/phosphate ratio and crystallinity of hydroxyapatite did not differ significantly among the three groups at either the diaphyseal or the metaphyseal regions. We next evaluated the collagen fiber orientation of cortical bone at the metaphysis and diaphysis of the femur using IR dichroism imaging. Collagen fibers were oriented in the direction of the bone axis in the diaphysis of the femur in Sham, OVX, and SKG mice. The orientation of the collagen, however, was inhomogeneous at the distal metaphysis of the femur in SKG mice compared with OVX and Sham mice (Fig. 4c). Quantitative analysis of the area ratio per unit, i.e., the area in which collagen fibers are well oriented to the direction of the bone axis (range, 1.5–3.0), revealed a significant decrease in the metaphysis of the femur in SKG mice compared with OVX and Sham mice (Fig. 4d). As for the degree of c-

Fig. 5 Arthritis-induced deterioration of mechanical properties in the periarticular bone adjacent to inflamed joints. a Comparisons of cortical thickness of the distal tibia around the breakdown point. b Comparisons of the mechanical properties of distal tibiae from the Sham, OVX, and SKG groups at 8 weeks after surgery or arthritis induction. SKG mouse

axis orientation of hydroxyapatite crystals in the bone matrix of cortical bone at the distal metaphysis of the tibia, which was measured by X-ray diffraction procedures, the degree did not differ significantly among the three groups (Fig. 4e). To better understand the mechanism responsible for the change in the periarticular bone quality induced by arthritis, we histologically analyzed the cortical bone at the metaphysis and diaphysis of the femur of SKG mice at 8 weeks after arthritis induction. Synovitis was observed at the distal metaphysis of the femur (Fig. 4f). The expression of DMP-1, a non-collagenous bone matrix protein produced specifically by osteocytes, was decreased at the distal metaphysis of the femur compared to the diaphysis of the femur (Fig. 4f).

Effects of arthritis on the biomechanical properties of periarticular bone We next evaluated the cortical bone geometry of the distal tibia using micro-CT before performing mechanical testing. Similarly to the distal femur, SKG mice exhibited a decrease in the Ct.Th of the distal tibia (Fig. 5a). The distal metaphyseal region of the tibia from SKG mice was significantly weaker and more ductile than that in the tibia from Sham and OVX mice. Compared to Sham mice, the distal tibia from SKG mice exhibited a significant decrease in terminal stress and toughness (Fig. 5b). The intrinsic material properties, including elastic modulus and bending strength, were decreased by nearly half in the SKG mice compared with the Sham mice, consistent with a previous report of the mechanical properties of the femur in

tibias exhibited significantly decreased ultimate stress and toughness compared with the Sham group, indicating an increase in the fragility fracture risk in the periarticular region adjacent to the inflamed joint. Values shown are means ± SD (*p < 0.05 vs Sham, †p < 0.05 vs OVX)

Osteoporos Int

SKG mice [22]. OVX had no significant effect on the biomechanical properties of the distal metaphysis of the tibia.

Discussion The present study demonstrated that autoimmune arthritis induces especially severe osteoporosis in the periarticular region adjacent to inflamed joints. Although significant trabecular bone loss in the distal metaphysis of the femur was observed in both the SKG mouse model of autoimmune arthritis and the OVX mouse model of postmenopausal osteoporosis, the extent of the bone loss was greater in SKG mice than in OVX mice. Bone strength and fracture toughness in the periarticular region of the tibia adjacent to inflamed ankle joints were also significantly decreased in SKG mice, suggesting that arthritis increases the fragility fracture risk near inflamed joints, which is frequently observed in patients with RA. Our findings revealed that the decreased bone strength and fracture toughness of the periarticular bone in this mouse model of RA might be due to the deterioration of bone material quality, such as a decrease in the degree of mineralization and disorientation of collagen fibers, in addition to impaired cortical bone geometry. Cortical bone geometry and matrix properties were not altered in the diaphysis of the femur in either SKG or OVX mice but were changed in the distal metaphysis of the femur in SKG mice. Bone remodeling is much slower in cortical bone than in trabecular bone [23]; thus, it is not surprising that the quality and quantity of cortical bone were not altered in OVX mice. By contrast, in the SKG mice, we observed deterioration of the cortical bone geometry and matrix properties in periarticular bone, supporting the idea that inflammation and immobilization resulting from joint pain and destruction markedly affect bone metabolism in periarticular bone. Additionally, suppression of the expression of DMP-1, critical for proper bone mineralization, in periarticular bone is thought to be an important mechanism of bone fragility apart from the previous report that focused on arthritis-induced osteoclast activation [24]. With regard to the effect of arthritis on the periarticular bone matrix quality, our FTIR study revealed that arthritis led to a reduction in the degree of calcification as well as to collagen fiber disorientation. Although previous studies of animal arthritis models reported that chronic arthritis directly induces both quantitative and qualitative bone disturbances, leading to compromised biomechanical properties [22], most of the studies focused on trabecular bone because changes in trabecular bone occur faster and are more obvious than those in cortical bone [23]. From a biomechanical standpoint, however, alterations in the cortical bone geometry and properties are of great interest because cortical bone has a greater mechanical contribution than trabecular bone [23]. Therefore, our observations of changes in the cortical bone geometry and

matrix properties of periarticular bone have an important clinical implication when considering the timing of initiating osteoporosis medication and when assessing the fragility fracture risk in patients with RA. Arthritis did not affect the degree of orientation of hydroxyapatite examined by X-ray diffraction analysis. Although some previous studies reported that collagen fiber orientation and the c-axis of hydroxyapatite are generally co-linear, Skedros et al. demonstrated that the dissociation of mineral and collagen orientation may differentially adapt compact bone to regional loading environments [25–27]. Therefore, changes in the hydroxyapatite orientation may occur after a longer period of time than changes in the collagen orientation caused by arthritis. This requires further investigation, however, because the FTIR spectrum was acquired from undecalcified tissue sections of the distal femur, while the Xray diffraction data on hydroxyapatite orientation were obtained from the distal metaphysis of the tibia. The present study has several limitations. First, we used the distal metaphysis of the tibia as periarticular bone adjacent to the inflamed joint for mechanical testing despite having performed micro-CT and FTIR analyses on the distal metaphysis of the femur, because the geometry of the distal metaphysis of the femur did not allow us to perform appropriate biomechanical testing to evaluate the mechanical properties of cortical bone, whereas the distal metaphysis of the tibia, having less trabecular bone, has a comparatively appropriate geometry for evaluating the mechanical properties involved in bending stress. In addition, given that ankle arthritis develops earlier and more severely than knee joint arthritis in SKG mice, the distal metaphysis of the tibia may be more severely affected by arthritis. Second, we compared immunized SKG mice with OVX and Sham-operated mice with a normal BALB/c background. Before we performed the OVX or induced arthritis, we evaluated the bone microstructure of both strains and confirmed that there were no differences between the two strains. Additionally, SKG mice might respond differently to OVX than BALB/c mice. Estrogen deficiency-induced bone loss is mainly due to the upregulated production of tumor necrosis factor-α, interleukin-1, and interleukin-6 by activated T cells [28, 29]. Given that the mutation of the ZAP-70 gene, which is a key signal transduction molecule in T cells, in SKG mice is reported to alter the production of proinflammatory cytokines, such as tumor necrosis factor-α, interleukin-1, and interleukin-6 [30], SKG mice might respond differently to OVX than wild-type mice. Therefore, we examined differences in the bone phenotype of immunized SKG mice and operated normal mice. Third, 12 weeks of age is too young to be a reasonable model of postmenopausal osteoporosis. Because the young age at which OVX was conducted may have lessened the severity of bone loss, it seems inappropriate to draw a conclusion as to whether OVX or arthritis has a greater impact on bone loss based on the results of this study.

Osteoporos Int

In conclusion, periarticular osteoporosis induced by autoimmune arthritis, which is not seen in OVX, is characterized by deterioration of the cortical bone geometry and quality, as well as bone quantity. These qualitative and quantitative changes lead to bone fragility in periarticular bone adjacent to inflamed joints. Acknowledgments This project was supported in part by a Grant-inAid for Scientific Research (C) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan 25462357 (M. Takahata) and by a Grant-in-Aid for Scientific Research (A), MEXT (No. 15H02207) (S. Tadano). Compliance with ethical standards The Ethics Review Committee for Animal Experimentation of Hokkaido University approved the experimental protocol of this study. Conflicts of interest None.

References 1.

Alamanos Y, Drosos AA (2005) Epidemiology of adult rheumatoid arthritis. Autoimmun Rev 4(3):130–136. doi:10.1016/j. autrev.2004.09.002 2. McInnes IB, Schett G (2011) The pathogenesis of rheumatoid arthritis. N Engl J Med 365(23):2205–2219. doi:10.1056/NEJMra1004965 3. van Staa TP, Geusens P, Bijlsma JW, Leufkens HG, Cooper C (2006) Clinical assessment of the long-term risk of fracture in patients with rheumatoid arthritis. Arthritis Rheum 54(10):3104– 3112. doi:10.1002/art.22117 4. Orstavik RE, Haugeberg G, Uhlig T, Falch JA, Halse JI, Hoiseth A, Lilleas F, Kvien TK (2003) Vertebral deformities in 229 female patients with rheumatoid arthritis: associations with clinical variables and bone mineral density. Arthritis Rheum 49(3):355–360. doi:10.1002/art.11118 5. Alenfeld FE, Diessel E, Brezger M, Sieper J, Felsenberg D, Braun J (2000) Detailed analyses of periarticular osteoporosis in rheumatoid arthritis. Osteoporos Int 11(5):400–407. doi:10.1007/s001980070106 6. Jensen TW, Hansen MS, Horslev-Petersen K, Hyldstrup L, Abrahamsen B, Langdahl B, Zerahn B, Podenphant J, Stengaard-Petersen K, Junker P, Ostergaard M, Lottenburger T, Ellingsen T, Andersen LS, Hansen I, Skjodt H, Pedersen JK, Lauridsen UB, Svendsen AJ, Tarp U, Lindegaard H, Jurik AG, Vestergaard A, Hetland ML (2014) Periarticular and generalised bone loss in patients with early rheumatoid arthritis: influence of alendronate and intra-articular glucocorticoid treatment. Post hoc analyses from the CIMESTRA trial. Ann Rheum Dis 73(6):1123–1129. doi:10.1136/annrheumdis-2012-203171 7. Rothschild B (2012) Comment on: Periarticular osteoporosis: a useful feature in the diagnosis of early rheumatoid arthritis? Reliability and validity in a cross-sectional diagnostic study using dual-energy X-ray absorptiometry. Rheumatology (Oxford) 51(7): 1340–1341. doi:10.1093/rheumatology/kes044, author reply 1341 8. Iwata T, Ito H, Furu M, Hashimoto M, Fujii T, Ishikawa M, Yamakawa N, Terao C, Azukizawa M, Hamamoto Y, Mimori T, Akiyama H, Matsuda S (2015) Periarticular osteoporosis of the forearm correlated with joint destruction and functional impairment in patients with rheumatoid arthritis. Osteoporos Int. doi:10.1007/s00198-015-3256-1 9. Jimenez-Boj E, Redlich K, Turk B, Hanslik-Schnabel B, Wanivenhaus A, Chott A, Smolen JS, Schett G (2005) Interaction

between synovial inflammatory tissue and bone marrow in rheumatoid arthritis. J Immunol 175(4):2579–2588 10. Shigeyama Y, Pap T, Kunzler P, Simmen BR, Gay RE, Gay S (2000) Expression of osteoclast differentiation factor in rheumatoid arthritis. Arthritis Rheum 43(11):2523–2530. doi:10.1002/15290131(200011)43:11<2523::AID-ANR20>3.0.CO;2-Z 11. Gravallese EM, Manning C, Tsay A, Naito A, Pan C, Amento E, Goldring SR (2000) Synovial tissue in rheumatoid arthritis is a source of osteoclast differentiation factor. Arthritis Rheum 43(2):250–258. doi:10.1002/1529-0131(200002)43:2<250::AID-ANR3>3.0.CO;2-P 12. Schett G, Gravallese E (2012) Bone erosion in rheumatoid arthritis: mechanisms, diagnosis and treatment. Nat Rev Rheumatol 8(11): 656–664. doi:10.1038/nrrheum.2012.153 13. Brandt KD (2003) Response of joint structures to inactivity and to reloading after immobilization. Arthritis Rheum 49(2):267–271. doi:10.1002/art.11009 14. Angeli A, Guglielmi G, Dovio A, Capelli G, de Feo D, Giannini S, Giorgino R, Moro L, Giustina A (2006) High prevalence of asymptomatic vertebral fractures in post-menopausal women receiving chronic glucocorticoid therapy: a cross-sectional outpatient study. Bone 39(2):253–259. doi:10.1016/j.bone.2006.02.005 15. Sakaguchi N, Takahashi T, Hata H, Nomura T, Tagami T, Yamazaki S, Sakihama T, Matsutani T, Negishi I, Nakatsuru S, Sakaguchi S (2003) Altered thymic T-cell selection due to a mutation of the ZAP-70 gene causes autoimmune arthritis in mice. Nature 426(6965):454–460. doi:10.1038/nature02119 16. Bouxsein ML, Boyd SK, Christiansen BA, Guldberg RE, Jepsen KJ, Muller R (2010) Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res 25(7):1468–1486. doi:10.1002/jbmr.141 17. Boskey A, Pleshko Camacho N (2007) FT-IR imaging of native and tissue-engineered bone and cartilage. Biomaterials 28(15):2465– 2478. doi:10.1016/j.biomaterials.2006.11.043 18. Hasegawa T, Amizuka N, Yamada T, Liu Z, Miyamoto Y, Yamamoto T, Sasaki M, Hongo H, Suzuki R, de Freitas PH, Oda K, Li M (2013) Sclerostin is differently immunolocalized in metaphyseal trabecules and cortical bones of mouse tibiae. Biomed Res 34(3):153–159 19. Giri B, Tadano S (2011) Nanostructural alteration in bone quantified in terms of orientation distribution of mineral crystals: a possible tool for fracture risk assessment. J Biomech Eng 133(12): 124503. doi:10.1115/1.4005482 20. Caetano-Lopes J, Nery AM, Henriques R, Canhao H, Duarte J, Amaral PM, Vale M, Moura RA, Pereira PA, Weinmann P, Abdulghani S, Souto-Carneiro M, Rego P, Monteiro J, Sakagushi S, Queiroz MV, Konttinen YT, Graca L, Vaz MF, Fonseca JE (2009) Chronic arthritis directly induces quantitative and qualitative bone disturbances leading to compromised biomechanical properties. Clin Exp Rheumatol 27(3):475–482 21. Keller KK, Thomsen JS, Stengaard-Pedersen K, Dagnaes-Hansen F, Nyengaard JR, Hauge EM (2012) Bone formation and resorption are both increased in experimental autoimmune arthritis. PLoS One 7(12):e53034. doi:10.1371/journal.pone.0053034 22. Caetano-Lopes J, Nery AM, Canhao H, Duarte J, Cascao R, Rodrigues A, Perpetuo IP, Abdulghani S, Amaral PM, Sakaguchi S, Konttinen YT, Graca L, Vaz MF, Fonseca JE (2010) Chronic arthritis leads to disturbances in the bone collagen network. Arthritis Res Ther 12(1):R9. doi:10.1186/ar2908 23. Bala Y, Zebaze R, Seeman E (2015) Role of cortical bone in bone fragility. Curr Opin Rheumatol 27(4):406–413. doi:10.1097/BOR.0000000000000183 24. Gough A, Sambrook P, Devlin J, Huissoon A, Njeh C, Robbins S, Nguyen T, Emery P (1998) Osteoclastic activation is the principal mechanism leading to secondary osteoporosis in rheumatoid arthritis. J Rheumatol 25(7):1282–1289

Osteoporos Int 25.

26.

27.

Skedros JG, Sorenson SM, Takano Y, Turner CH (2006) Dissociation of mineral and collagen orientations may differentially adapt compact bone for regional loading environments: results from acoustic velocity measurements in deer calcanei. Bone 39(1):143–151. doi:10.1016/j.bone.2005.12.007 Reisinger AG, Pahr DH, Zysset PK (2011) Principal stiffness orientation and degree of anisotropy of human osteons based on nanoindentation in three distinct planes. J Mech Behav Biomed Mater 4(8):2113–2127. doi:10.1016/j. jmbbm.2011.07.010 Yamada S, Tadano S, Fujisaki K, Kodaki Y (2013) Influence of osteon area fraction and degree of orientation of HAp crystals on mechanical properties in bovine femur. J Biomech 46(1):31–35. doi:10.1016/j.jbiomech.2012.09.020

28.

Roggia C, Gao Y, Cenci S, Weitzmann MN, Toraldo G, Isaia G, Pacifici R (2001) Up-regulation of TNF-producing T cells in the bone marrow: a key mechanism by which estrogen deficiency induces bone loss in vivo. Proc Natl Acad Sci U S A 98(24):13960– 13965. doi:10.1073/pnas.251534698 29. Ammann P, Rizzoli R, Bonjour JP, Bourrin S, Meyer JM, Vassalli P, Garcia I (1997) Transgenic mice expressing soluble tumor necrosis factor-receptor are protected against bone loss caused by estrogen deficiency. J Clin Invest 99(7):1699–1703. doi:10.1172/JCI119333 30. Hata H, Sakaguchi N, Yoshitomi H, Iwakura Y, Sekikawa K, Azuma Y, Kanai C, Moriizumi E, Nomura T, Nakamura T, Sakaguchi S (2004) Distinct contribution of IL-6, TNF-alpha, IL1, and IL-10 to T cell-mediated spontaneous autoimmune arthritis in mice. J Clin Invest 114(4):582–588. doi:10.1172/JCI21795