Plasminogen activator inhibitor-1 deficiency enhances subchondral osteopenia after induction of osteoarthritis in mice

Macrophage colony-stimulating factor (M-CSF) and receptor activator of nuclear factor-κB ligand (RANKL) were purchased from Wako (Osaka, Japan). SB431542, interleukin-1β (IL-1β) and transforming growth factor-β₁ (TGF-β₁) were obtained from Tocris (Bristol, UK), R&D systems (Minneapolis, MN, USA) and Sigma (St Louis, MO, USA), respectively. Bone morphogenetic protein-2 (BMP-2) was obtained from Pfizer Inc. (Groton, CT, USA).

PAI-1 KO and counterpart WT mice were used in this study. WT and PAI-1 KO mice were kindly provided by D. Collen (University of Leuven, Leuven, Belgium) and were bred at Kindai University Faculty of Medicine animal facility. The genetic background of all mice was mixed C57BL/6 J (81.25%) and 129/SvJ (18.75%). Female WT and PAI-1 KO mice were randomly divided into four groups: WT/Sham (n = 11), WT/ovariectomy (OVX) (n = 11), PAI-1 KO/Sham (n = 10) and PAI-1 KO/OVX (n = 11). Seven-week-old female WT and PAI-1 KO mice were ovariectomized or sham-operated as previously described [16]. One week after the OVX or sham surgery, all mice received destabilization of the medial meniscus (DMM) surgery on the right knee due to induction of OA and sham surgery on the left knee. Eight weeks after the DMM surgery, these mice were subjected to analysis. Food and water were provided ad libitum. The room temperature was kept at 24 ± 1 °C with a 12 h:12 h light/dark cycle. All experiments were performed according to the guidelines of the National Institutes of Health and the institutional rules for the use and care of laboratory animals at Kindai University.

The DMM model, which provided high reproducibility and a slower progression of OA in mice, was used for evaluation of changes in subchondral bone and cartilage in the OA state in WT and PAI-1 KO mice. DMM surgery was performed on the right knee of mice as previously described [17]. Briefly, under anesthesia induced by 2% isoflurane, the anterior skin was cut by 5 mm to expose the joint capsule in mice. To expose the medial meniscus, the joint capsule immediately medial to the patellar tendon was incised. After transection of the meniscotibial ligament, skin was closed sterilely. A sham surgery was performed on the left knee using the same procedure with no transection of the meniscotibial ligament.

For qCT analysis of the bone mineral density (BMD), mice were anesthetized using 2% isoflurane and scanned using an experimental animal CT system (LaTheta LCT-200, Hitachi Aloka Medical, Tokyo, Japan) [18]. CT scans were performed at a tube voltage of 50 kVp, a tube current of 500 μA, an integration time of 3.6 ms, an axial field of view of 48 mm, and an isotopic voxel size of 24 μm. To assess the trabecular BMD of the tibial subchondral bone, regions of interest (ROI) were defined as 240-μm (10-slice) segments from 72 μm proximal to the end of the proximal growth plate towards the joint. This ROI was 200 μm distal towards the joint, compared to ROI for the assessment of the trabecular BMD for systemic osteopenia in tibia. An increase in bone mineral content occurs in the most proximate subchondral region of the cartilage tissues in the knee. We therefore employed the most proximate trabecular bone region with a decrease in BMD with DMM and OVX in our preliminary experiments as the ROI for the measurement of BMD of the subchondral bone. Bone parameters were analyzed using LaTheta software (version3.40). A threshold density of 160 mg/cm³ was set to distinguish mineralized from unmineralized tissue. The density range was calibrated daily with a phantom supplied by the manufacturer.

Mice were euthanized with excess isoflurane 8 weeks after DMM surgery. The knee samples were removed, fixed in 4% paraformaldehyde for 24 h, and decalcified in 22.5% formic acid and 340 mM sodium citrate solution for 24 h. After demineralization, specimens were embedded in paraffin and cut into 5-μm sections. The sections were stained with hematoxylin/eosin and photographed under a microscope (BZ-9000, Keyence, Osaka, Japan). The sections were stained with Safranin-O and fast green, and subsequently counterstained with hematoxylin. Each knee was evaluated for the progression of arthropathy with 6 grades using the Osteoarthritis Research Society International (OARSI) histological scoring system [19].

Total RNA was extracted from primary osteoblasts using the RNeasy Mini Kit (Qiagen, Tokyo, Japan). One μg total RNA was reverse transcribed using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster, CA, USA). Quantitative real-time PCR was performed using StepOnePlus and the Fast SYBR Green PCR Master Mix (Life Technologies, Tokyo, Japan) as previously described [13]. A list of primers used is shown in Table 1. The specific mRNA amplification of the target was determined as the Ct value, which was followed by normalization by the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) mRNA level.

Primary osteoblastic cells were obtained from the calvaria of 3-day-old female WT and PAI-1 KO mice, as previously described [20]. The calvaria was removed from soft tissue, and digested 4 times with 1 mg/mL collagenase and 0.25% trypsin for 20 min at 37 °C. Cells from second, third and fourth digestions were collected and grown in Minimum Essential Medium Alpha Modification (α-MEM; Wako) with 10% fetal bovine serum (FBS; HyClone, Logan, UT, USA). The medium was changed twice a week.

Bone marrow stromal cells were obtained from female WT and PAI-1 KO mice as previously described [20]. Briefly, femur and tibia were removed and subsequently the bone marrow cells were flushed out into Dulbecco’s Modified Eagle’s Medium (DMEM; Wako) with 10% FBS. The nonadherent cells were removed by washing using phosphate-buffered saline.

Bone marrow cells were collected from the femur and tibia of 8-week-old female WT and PAI-1 KO mice. Analysis of osteoclast formation was performed as previously reported [21]. Briefly, bone marrow cells were plated in a 24-well plate (5.0 × 10⁵ cells/well) and cultured in α-MEM with 10% FBS and 50 ng/mL M-CSF for 3 days at 37 °C. Then, osteoclasts were formed in α-MEM with 10% FBS, 50 ng/mL M-CSF and 50 ng/mL RANKL for a further 3 days at 37 °C. Detection of osteoclasts was performed using a tartrate-resistant acid phosphatase (TRAP) staining kit (Wako). The numbers of TRAP-positive multinucleated cells (MNCs) were counted in a certain area within the well.

We evaluated the trabecular BMD of the tibial subchondral bone, in which ROI were defined as 240 μm (10-slice) segments from 72 μm proximal to the end of the proximal growth plate towards the joint using quantitative CT analysis. OVX significantly decreased the trabecular BMD of the subchondral bone in WT and PAI-1 KO mice (Fig. 1a, b). DMM significantly decreased the trabecular BMD of the subchondral bone in PAI-1 KO mice, although DMM did not affect it in WT mice with or without OVX. PAI-1 deficiency significantly augmented a decrease in the trabecular BMD of the subchondral bone in mice with OVX and DMM (Fig. 1a, b). Histological analysis findings were compatible with those results in qCT (Fig. 1c).

Next, we examined the effects of OVX, DMM and PAI-1 deficiency on articular cartilage degeneration in mice. The progression of arthropathy at the knees was evaluated by an OARSI scoring system, which is a widely used method for the evaluation of osteoarthritis. As shown in Fig. 2, there were no significant differences in the OARSI score between WT and PAI-1 KO mice with or without OVX. DMM significantly increased the OARSI score in WT and PAI-1 KO mice, although OVX did not affect the OARSI score. OARSI scores seemed to be elevated in PAI-1 KO mice, compared to those in WT mice, especially in mice with both OVX and DMM, although their differences were not statistically significant.

We investigated the mechanisms by which PAI-1 deficiency augments subchondral osteopenia induced by DMM in mice using mouse primary osteoblasts from WT and PAI-1 KO mice. IL-1β and TGF-β are important and representative factors for the pathogenesis of OA [22]. We therefore examined the roles of IL-1β and TGF-β signals in WT and PAI-1 KO mice to clarify the roles of PAI-1 in the pathological state of OA in vitro. Both IL-1β and TGF-β significantly enhanced PAI-1 mRNA levels in mouse osteoblasts (Fig. 3a). However, IL-1β did not affect TGF-β mRNA levels in these cells from both WT and PAI-1 KO mice (Fig. 3b). Next, we examined the effects of IL-1β and PAI-1 deficiency on osteoblast differentiation in mouse osteoblasts from WT and PAI-1 KO mice. As shown in Fig. 4a, IL-1β significantly decreased Osterix mRNA levels in WT mice, although its effects on other osteogenic genes, such as Runx2, alkaline phosphatase (ALP), type I collagen (Col-1) and osteocalcin (OCN) were not significant. PAI-1 deficiency did not affect Osterix mRNA levels suppressed by IL-1β (Fig. 4a). Moreover, SB431542, an inhibitor of activin-like kinase (ALK)5, TGF-β receptor type I kinase, did not affect Osterix mRNA levels suppressed by IL-1β, although it significantly blunted Osterix mRNA levels suppressed by TGF-β in osteoblasts (Fig. 4b). We next investigated the involvement of PAI-1 in osteoblast differentiation from mouse bone marrow cells in the presence or absence of IL-1β. As shown in Fig. 4c, PAI-1 deficiency did not affect the mRNA levels of Osterix, ALP and OCN in bone marrow cells treated with BMP-2 in the presence or absence of IL-1β, suggesting that PAI-1 deficiency does not affect osteoblastic differentiation from mouse bone marrow cells.

Finally, we investigated the effects of PAI-1 deficiency on osteoclast formation in WT and PAI-1 KO mice. RANKL is a crucial factor for osteoclast formation and activity, and osteoprotegerin (OPG) inhibits osteoclast formation by binding to RANKL. RANKL and OPG are the chief regulators of bone resorption. As shown in Fig. 5, IL-1β significantly elevated RANKL mRNA levels as well as the ratio of RANKL/OPG mRNA in osteoblasts from WT mice. PAI-1 deficiency significantly augmented this IL-1β-enhanced RANKL mRNA level and RANKL/OPG mRNA ratio in osteoblasts. However, SB431542, a TGF-β signaling inhibitor, did not affect the IL-1β-enhanced RANKL mRNA level and RANKL/OPG mRNA ratio in osteoblasts from mice with or without PAI-1 deficiency.

We examined the effects of PAI-1 deficiency of osteoclast formation from mouse bone marrow cells obtained from WT and PAI-1 KO mice. PAI-1 deficiency significantly increased the number of TRAP-positive MNCs in the presence or absence of IL-1β in mouse bone marrow cell cultures in the presence of M-CSF and RANKL (Fig. 6).