Takanobu Nakase

Mechanical Tension-Stress Induces Expression of Bone Morphogenetic Protein (BMP)-2 and BMP-4, but Not BMP-6, BMP-7, and GDF-5 mRNA, During Distraction Osteogenesis

Bone lengthening with osteotomy and gradual distraction was achieved in 57 rats, and the effect of mechanical tension‐stress on gene expression of bone morphogenetic proteins (BMPs) was investigated by in situ hybridization and Northern blot analysis using probes of BMP‐2, BMP‐4, BMP‐6, BMP‐7, and growth/differentiation factor (GDF)‐5. There was a lag phase for 7 days after femoral osteotomy until gradual distraction was carried out for 21 days at a rate of 0.25 mm/12 h using a small external fixator. The signals of the above BMPs mRNA were not detected in the intact rat bone but they were induced after osteotomy except those for BMP‐7. By 4 days after osteotomy, BMP‐2 and BMP‐4 mRNAs were detected in chondrogenic precursor cells in the subperiosteal immature callus. BMP‐6 and GDF‐5 mRNA were detected in more differentiated cells in chondroid bone. By 7 days after osteotomy, cartilaginous external callus and bony endosteal callus were formed. Meanwhile, the signals of BMP‐2 and BMP‐4 mRNAs declined to preoperative levels, whereas the signals of BMP‐6 and GDF‐5 mRNAs were rather elevated. As distraction was started, the callus elongated and eventually separated into proximal and distal segments forming a fibrous interzone in the middle. Expression of BMP‐2 and BMP‐4 mRNAs was markedly induced at this stage. Their signals were detected widely among chondrogenic and osteogenic cells and their precursor cells sustaining mechanical tension‐stress at the fibrous interzone. BMP‐6 and GDF‐5 mRNAs were detected exclusively in chondrogenic cells at both ends of the fibrous interzone, where endochondral ossification occurred. But neither mRNA was detected in terminally differentiated hypertrophic chondrocytes. As distraction advanced, the cartilage was progressively resorbed from both ends and new bone was formed directly by intramembranous ossification. There was no new cartilage formation in the advanced stage of distraction. The signals of BMP‐6 and GDF‐5 mRNA declined by this stage, while those of BMP‐2 and BMP‐4 were maintained at high level for as long as distraction was continued. After completion of distraction, the fibrous interzone fused and the lengthened segment was consolidated. BMP‐2, BMP‐4, BMP‐6, nor GDF‐5 was expressed at this stage. The signals of BMP‐7 were not detected throughout the experiment. The present results suggest that excellent and uninterrupted bone formation during distraction osteogenesis owes to enhanced expression of BMP‐2 and BMP‐4 genes by mechanical tension‐stress. Abundant gene products of BMP‐2 and BMP‐4 could induce in situ bone formation by paracrine and autocrine mechanisms.

Expression of Bone Matrix Proteins mRNA During Distraction Osteogenesis

Distraction osteogenesis is a recently advanced principle of bone lengthening in which a bone separated by osteotomy is subjected to slow progressive distraction using an external fixation device. Appropriate mechanical tension‐stress is believed not to break the callus but rather to stimulate osteogenesis. To study the molecular features of this process, the expression and localization of the mRNAs encoding osteopontin (OPN), osteocalcin (OC), matrix Gla protein (MGP), osteonectin (ON), and collagen type I and II during distraction osteogenesis were examined by in situ hybridization and Northern blot analysis. The process can be divided into three distinct phases: the lag phase for 7 days between osteotomy and the beginning of distraction, the distraction phase for 21 days, and the consolidation phase for several weeks. The histologic and molecular events taking place during the lag phase were similar to those observed in fracture healing. The osteotomy site was surrounded by external callus consisting of hyaline cartilage. As distraction started at the rate of 0.25 mm/12 h, the cartilaginous callus was elongated, deformed, and eventually separated into proximal and distal segments. The chondrocytes were stretched along the tension vector and became fibroblast‐like in shape. Although morphologically these cells were distinguishable from osteogenic cells, they expressed OPN, OC, and alkaline phosphatase mRNAs. As distraction advanced, the cartilaginous callus was progressively replaced by bony callus by endochondral ossification and thereafter new bone was formed directly by intramembranous ossification. OPN mRNA was detected in preosteoblasts and osteoblasts at the boundary between fibrous tissue and new bone. ON, MGP, and OC mRNAs appeared early in the differentiation stage. The variety of cell types expressing mRNA encoding bone matrix proteins in distraction osteogenesis was much greater than that detected in the embryonic bone formation and fracture healing process. Moreover, the levels of OPN, ON, MGP, and OC mRNA expression markedly increased during the distraction phase. These results suggested that mechanical tension‐stress modulates cell shape and phenotype, and stimulates the expression of the mRNA for bone matrix proteins.

Regulation of proliferation and osteochondrogenic differentiation of periosteum-derived cells by transforming growth factor-beta and basic fibroblast growth factor.

We studied the effects of transforming growth factor-beta and basic fibroblast growth factor on the regulation of proliferation and osteochondrogenic differentiation of periosteum-derived cells, which have the potential to differentiate into bone and hypertrophic cartilage in vitro. Histological observation revealed that transforming growth factor-beta stimulated chondrogenesis of periosteum-derived cells while basic fibroblast growth factor stimulated proliferation of fibroblast-like cells and inhibited osteochondrogenic differentiation. Immunohistochemical studies revealed that basic fibroblast growth factor inhibited the expression of osteocalcin. Transforming growth factor-beta enhanced uronic acid content but decreased DNA content, alkaline phosphatase activity, and calcium content. In contrast, basic fibroblast growth factor enhanced DNA content but decreased alkaline phosphatase activity, calcium content, and uronic acid content. In addition, transforming growth factor-beta shortened the time-course of gene expression of type-X collagen whereas basic fibroblast growth factor inhibited the gene expression. These results indicate that transforming growth factor-beta stimulates osteochondrogenic differentiation of periosteum-derived cells but inhibits proliferation. They also indicate that basic fibroblast growth factor stimulates proliferation of periosteum-derived cells but inhibits osteochondrogenic differentiation.

Bone Morphogenetic Protein Signals Are Required for Cartilage Formation and Differently Regulate Joint Development During Skeletogenesis

The bone morphogenetic protein (BMP) family consists of a large number of members and has diverse biological activities during development. Various tissues express pleural BMP family members, which seem to cooperatively regulate developmental events. Here, multiple BMP signals were inactivated in chondrocytes to clarify the function of BMPs during skeletogenesis. To obtain tissue‐specific inactivation, Noggin gene (Nog) was overexpressed in cartilage under the control of α2(XI) collagen gene (Col11a2) promoter/enhancer sequences. The resultant transgenic mice lacked most of their cartilaginous components, suggesting that cartilage does not develop without BMP signals. These effects seem to be mediated through down‐regulation of Sox9 expression. Conversely, specific BMP signals were activated in the skeleton by targeted expression of Bmp4 in cartilage and the resultant phenotype was compared with that of transgenic mice expressing growth and differentiation factor‐5 (GDF‐5), another BMP family member. Overactivity of Bmp4 in the skeleton caused an increase of cartilage production and enhanced chondrocyte differentiation, as GDF5 expression did, but it did not disturb joint formation as GDF5 did. During skeletogenesis, unique roles of each BMP may reside in the regulation of joint development. Together with the common effect on the cartilage overproduction by Bmp4 and GDF5 overactivation, loss of cartilage by inactivation of multiple BMPs in Noggin transgenic mice indicates that signals for cartilage production are reinforced by multiple BMPs exclusively. These conclusions may account for the reason why multiple BMPs are coexpressed in cartilage.

Alterations in the expression of osteonectin, osteopontin and osteocalcin mRNAs during the development of skeletal tissues in vivo

Heterogeneity in the expression of three members of non-collagenous matrix proteins in osteogenic and chondrogenic development in vivo were investigated by in situ hybridization. Sections of several skeletal tissues from mice at various stages of development were hybridized with digoxigenin-labeled complementary RNA probes encoding osteonectin (Osn), osteopontin (Osp) and osteocalcin (Osc). In calvariae and mandibulae, Osn messenger RNA (mRNA) was detected in cells in pre-osseous and osseous tissues before mineralization. Osp mRNA was found in cells attached to the mineralized bone matrix together with Osn mRNA followed by the expression Osc mRNA. In long bones, mRNAs for Osn, Osp and Osc were sequentially expressed with bone development from primary spongiosa to diaphyseal bone. In growth cartilage, Osn mRNA was observed in chondrocytes in non-mineralized cartilage, whereas Osp mRNA was detected in hypertrophic chondrocytes in mineralized cartilage matrix with a characteristic switch in expression. Osc mRNA was not detected in any chondrocytes. These results indicate that osteogenic differentiation in bone development in vivo is characterized by the sequential expression of these three genes, and suggest that these genes are expressed differentially and specifically, in association with extra-cellular matrix mineralization.

The relationship between apoptosis of endplate chondrocytes and aging and degeneration of the intervertebral disc.

Study Design. Apoptosis in cervical intervertebral disc cells and cartilaginous endplate cells was examined by the nick end labeling (TUNEL) technique during the process of natural aging and in a mouse experimental spondylosis model. Objectives. To determine the role of apoptosis in aging and degeneration of intervertebral discs by monitoring chronologic changes in the quantity and localization of apoptotic cells. Summary of Background Data. Apoptosis occurs within human intervertebral discs, but little is known about the pathologic significance of this process. On the other hand, the cartilaginous endplate is known to decrease in thickness and to disappear with aging and degeneration. The cause of this age-related change remains unclear. Methods. A mouse spondylosis model was prepared via surgical resection of the posterior spinal element in 12 mice to examine the experimentally induced spondylosis process. Eighteen naturally aged mice were also used to examine the influence of aging. Paraffin-embedded midsagittal sections of the cervical spine were obtained 2, 3, 6, and 12 months after surgery in the spondylosis model and in the age-matched naturally aged mice, as well as in 4-week-old and 18-month-old naturally aged mice. Sections were stained with hematoxylin and eosin, safranin-O, and the TUNEL procedure. The number of apoptotic cells and vital cells were counted in the cartilaginous endplate of the intervertebral disc excluding the growth cartilage, and the degree of disappearance of the cartilaginous endplate was evaluated. Results. Apoptosis, particularly noticeable in the cartilaginous endplate, increased with age and resulted in a marked decrease in cell density. Subsequently, the structure of the cartilaginous endplate began to disappear. Apoptosis was more evident and the structure of the cartilaginous endplate began to disappear more rapidly in the surgically treated group than in the naturally aged group. Conclusions. TUNEL-positive cells in the cartilaginous endplate increased with age, with destruction of the cartilaginous endplate after apoptosis (TUNEL-positive cell death). The application of the spondylosis model increased the incidence of apoptosis preceding the development of spondylosis. This suggests that apoptosis plays a role in the age-related changes seen in the cartilaginous endplate of the intervertebral disc and in the experimentally induced spondylosis process.

Interleukin-1β enhance and tumor necrosis factor-α inhibits bone morphogenetic protein-2-induced alkaline phosphatase activity in MC3T3-E1 osteoblastic cells

The modulatory effects of interleukin (IL)-1β and tumor necrosis factor (TNF)-α on bone morphogenetic protein (BMP)-2- and -4-induced alkaline phosphatase (ALP) activity were examined in cultures of mouse MC3T3-E1 osteoblastic cells. Both BMP-2 and -4 significantly induced ALP in these cells. IL-1β alone had no effect on ALP activity, but it significantly enhanced BMP-2- and -4-induced ALP activity. TNF-α suppressed the induction of ALP by BMP-2 or -4. The results suggest that the action of BMP on osteogenic differentiation may be regulated by such immuno/inflammatory cytokines as IL-1β and TNF-α.

Localization of bone morphogenetic protein-2 in human osteoarthritic cartilage and osteophyte

OBJECTIVES To examine the localization of bone morphogenetic protein (BMP)-2 mRNA and protein in human osteoarthritic (OA) articular cartilage and osteophyte. DESIGN Five normal, four growing and 14 OA human cartilage samples, graded histomorphologically by Mankin Score, were studied by in situ hybridization and immunohistochemistry for the expression of BMP-2. RESULTS BMP-2 mRNA was present in chondrocytes in neonatal growing articular cartilage, but was scarcely present in normal adult articular cartilage. In OA articular cartilage, BMP-2 mRNA and protein were detected in both clustering and individual chondrocytes in moderately or severely damaged OA cartilage. In moderately damaged OA cartilage, BMP-2 mRNA was localized in both upper and middle zone chondrocytes, but was not detected in deep layer chondrocytes. In severely damaged OA cartilage, cellular localization of BMP-2 mRNA was extended to the deep zone. In the area of osteophyte formation, BMP-2 mRNA was intensely localized in fibroblastic mesenchymal cells, fibrochondrocytes, chondrocytes and osteoblasts in newly formed osteophytic tissue. The pattern of BMP-2/4 immunolocalization was associated with that of mRNA localization. CONCLUSIONS BMP-2 mRNA and BMP-2/4 were detected in cells appearing in OA tissues. BMP-2 was localized in cells of degenerating cartilage as well as osteophytic tissue. Given the negative localization of BMP-2 in normal adult articular cartilage, BMP-2 might be involved in the regenerating and anabolic activities of OA cells, which respond to cartilage damage occurring in osteoarthritis.

Mechanisms of age-related decline in insulin-like growth factor-I dependent proteoglycan synthesis in rat intervertebral disc cells.

Study Design. Age-related fluctuations in insulin-like growth factor-I dependent proteoglycan synthesis in rat intervertebral disc cells were investigated. Objectives. The purpose of this study was to determine whether synthetic responses to insulin-like growth factor-I decline with age and to explore the possibility that an age-related increase in the expression of insulin-like growth factor binding proteins suppresses matrix synthesis in intervertebral disc cells. Summary and Background Data. Several studies have reported that the responsiveness of chondrocytes to insulin-like growth factor-I decreases with age and furthermore that this phenomenon may be related to increased expression of insulin-like growth factor binding proteins by chondrocytes. Materials and Methods. Nucleus pulposus tissue and cells were obtained from the coccygeal vertebrae of 8-week-old, 40-week-old, and 120-week-old rats. Age-related changes in the expression of insulin-like growth factor-I and its receptor were assessed together with insulin-like growth factor-I dependent proteoglycan synthesis by the cultured nucleus pulposus cells. Also, western blot analysis of insulin-like growth factor binding protein-1 was carried out, and further examination was performed of insulin-like growth factor-I signal transduction through tyrosine phosphorylation of insulin receptor substrate-1, which is a signal transducer of insulin-like growth factor-I. Results. Semiquantitative reverse transcription polymerase chain reaction analysis indicated that the expression of insulin-like growth factor-I receptor in 120-week cells decreased clearly in comparison with the cells of younger animals. By contrast, insulin-like growth factor-I dependent proteoglycan synthesis decreased with age, and the sharpest decline of synthesis was found between 8-week and 40-week cells, although the level of insulin-like growth factor-I/insulin-like growth factor-I receptor gene expression was maintained in 40-week-old animals. Consistent with the results of proteoglycan synthesis, the expression of phosphorylated insulin receptor substrate-1 decreased with age. Thus, the expression of insulin-like growth factor binding protein-1 and proteoglycan synthesis was investigated by use of Long R3 insulin-like growth factor-I, which was not influenced by insulin-like growth factor binding proteins. Insulin-like growthfactor binding protein-1 was strongly expressed in 40-week cells in comparison withthe expression in 8-week cells. Furthermore, proteoglycan synthesis in 40-week cells supplemented with Long R3 insulin-like growth factor-I was upregulated in comparison with that in 40-week cells supplemented with insulin-like growth factor-I. Conclusion. The present findings indicate that the age-related decline in insulin-like growth factor-I dependent proteoglycan synthesis in nucleus pulposus is caused, at least in part, by the increase in insulin-like growth factor binding proteins at the early stages of aging, and further suggest that a loss of proteoglycan synthesis during the late stages of aging is caused by the downregulation of insulin-like growth factor-I receptor in addition to an increase in insulin-like growth factor binding proteins.

Cervical lesions related to the systemic progression in rheumatoid arthritis

Study Design. Cross‐sectional study of cervical involvement in rheumatoid arthritis. Objectives. To clarify the correlation between the deterioration of cervical lesions and the systemic progression of rheumatoid arthritis. Summary of Background Data. The natural course of cervical lesions varies. To date, no systemic parameter has been clarified to predict the progression. Methods. One hundred seventy‐three patients with rheumatoid arthritis participated in this study. The authors studied the progression of cervical lesions and investigated the relation between the types of cervical subluxation at the end of study and the following four variables: the serum level of C‐reactive protein, the number of joints with erosion, carpal height ratio, and disease subset (least erosive subset, more erosive subset, and mutilating disease subset). Results. Of the 173 patients, 55 already had cervical subluxation before entering the study. During the follow‐up period, 44 patients deteriorated radiographically, and 77 (45%) had cervical involvement, including involvement of upper cervical lesions in 65 patients, upper lesions combined with subaxial subluxation in 10, and subaxial subluxation alone in 2. The upper cervical subluxation progressed in the order of anterior atlantoaxial subluxation, atlantoaxial subluxation combined with vertical subluxation, and vertical subluxation alone. Deterioration of upper cervical lesion and occurrence of subaxial subluxation were closely correlated with an elevation of serum C‐reactive protein level, an increase in the number of joints with erosion, and a decrease in the carpal height ratio. The incidence of cervical involvement and the extent of deterioration were different among the disease subsets. Conclusions. The serum level of C‐reactive protein, the number of joints with erosion, and the carpal height ratio correlated closely with the extent of the cervical subluxation. The average C‐reactive protein values during the follow‐up period correlated with progression of the cervical lesions. The classification of rheumatoid disease subset was useful for predicting the terminal feature of the cervical lesions.