help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Harada, H.
Right arrow Articles by Iijima, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harada, H.
Right arrow Articles by Iijima, T.
Right arrowPubmed/NCBI databases
*Substance via MeSH
Endocrinology Vol. 139, No. 9 3967-3975
Copyright © 1998 by The Endocrine Society

ARTICLES

Involvement of Lymphocyte Function-Associated Antigen-1 and Intercellular Adhesion Molecule-1 in Osteoclastogenesis: A Possible Role in Direct Interaction between Osteoclast Precursors

Hiroshi Harada, Toshio Kukita, Akiko Kukita, Yukihide Iwamoto and Tadahiko Iijima

Second Department of Oral Anatomy (H.H., T.K., T.I.), Faculty of Dentistry, Kyushu University, Fukuoka 812, Japan; Department of Orthopedics (H.H., Y.I.), Faculty of Medicine, Kyushu University, Fukuoka 812, Japan; and Department of Microbiology (A.K.), Saga Medical School, Saga 849, Japan

Address all correspondence and requests for reprints to: Toshio Kukita, Ph.D., Second Department of Oral Anatomy, Faculty of Dentistry, Kyushu University, 3–1-1 Maidashi, Fukuoka 812, Japan.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
In our search for molecules involved in the process of osteoclast differentiation, we examined the surface phenotypes of the preosteoclast-like cells and osteoclast-like multinucleated cells (MNCs) formed in bone marrow cultures, using monoclonal antibodies recognizing different antigen molecules expressed on hematopoietic cells. Among these cell surface antigens, lymphocyte function-associated antigen-1 (LFA-1) and intercellular adhesion molecule-1 (ICAM-1) were highly expressed on mononuclear cells in the cultures for forming preosteoclast-like mononuclear cells. The double detection of these two antigen molecules with osteoclast-specific antigen and with calcitonin receptor, using a fluorescence-activated cell sorter or autoradiography technique, revealed that LFA-1 and ICAM-1 were expressed on the preosteoclasts. The expression of ICAM-1 was detected on both preosteoclasts and osteoclast-like MNCs, whereas the expression of LFA-1 was restricted to preosteoclasts. We designed a peptide with the sequence of the binding site of ICAM-1 against the ligand LFA-1. In the whole bone marrow culture system for forming osteoclast-like MNCs, a significant inhibition of MNC formation was observed by the addition of this peptide. These results strongly suggest the involvement of an LFA-1/ICAM-1-interaction in osteoclastogenesis.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
OSTEOCLASTS are unique multinucleated cells (MNCs) responsible for bone resorption, and they are believed to be derived from a hematopoietic stem cell population. Osteoclast progenitors, recruited from the hematopoietic tissue to the bone tissue, proliferate and differentiate into the mononuclear precursors of osteoclasts, i.e. preosteoclasts, followed by fusion with each other to form multinucleated osteoclasts (1, 2). One approach for clarifying the molecular events concerning the osteoclastogenesis is to define the cell surface molecules involved in the process of osteoclast differentiation. Several research groups have attempted to identify the cell surface molecules expressed on preosteoclasts involved in the interaction with osteoblasts or in the specific recognition and specific fusion among preosteoclasts. The vascular cell adhesion molecule-1 (VCAM-1) expressed on osteoblasts participates in the interaction between osteoblasts and cells in the osteoclast lineage in the process of the differentiation of the osteoclasts (3, 4, 5). Kurachi et al. (6) reported the involvement of the intercellular adhesion molecule-1 (ICAM-1) in the process of the osteoclast formation. Mbalaviele et al. (7) showed the participation of E-cadherin in this fusion process. Kumegawa et al. (8) demonstrated the dynamic fusion process of preosteoclasts using time-lapse cinematography. Although this dynamic process has attracted the interest of many researchers, the precise molecular mechanism involved in the process of the recognition and fusion between preosteoclasts remains unknown.

We have developed a culture system in which the induction of mononuclear preosteoclast-like cells and the subsequent formation of osteoclast-like MNCs can be separated (9). In this culture system, mononuclear cells with characteristics of preosteoclasts are formed from the stromal cell-depleted bone marrow cells in the presence of heat-treated conditioned media of ROS17/2.8 cells. In addition, we have identified the lineage-specific cell surface antigen, designated Kat1 antigen, expressed on rat osteoclast (10, 11). This antigen is a reliable immunological marker of cells in osteoclast-lineage and is recognized by the anti-Kat1-antigen monoclonal antibody Kat1 (mAb Kat1). In the present study, using the culture system of osteoclastogenesis and the lineage-specific mAb Kat1, we analyzed the detailed surface phenotype of the preosteoclasts with the use of a fluorescence-activated cell sorter (FACS). The double detection of the calcitonin receptor (CTR) and cell surface antigens was also performed. Interestingly, lymphocyte function-associated antigen-1 (LFA-1) and ICAM-1 were found to be expressed on the preosteoclast-like mononuclear cells. We then estimated the involvement of the LFA-1/ICAM-1 interaction in osteoclastogenesis using an ICAM-1 peptide fragment (LFA-1/ICAM-1-blocking peptide) that corresponds to the domain of ICAM-1 and interacts with the ligand LFA-1. We demonstrated a suppressive effect of this peptide fragment on the formation of osteoclast-like MNCs.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
Male Sprague-Dawley rats (4–6 weeks old) were obtained from SEAC Yoshitomi (Fukuoka, Japan). Both {alpha}MEM and FCS were purchased from Gibco (Grand Island, NY). 1{alpha},25-dihydroxyvitamin D3 (1{alpha},25(OH)2D3) was purchased from Biomol (Plymouth Meeting, PA). The rat osteosarcoma cell line ROS 17/2.8 was kindly provided by Dr. L. Bonewald (University of Texas Health Science Center, San Antonio, TX). A commercial kit (Sigma Chemicals, St. Louis, MO) was used for the tartrate-resistant acid phosphatase (TRAP) staining. 125I-labeled salmon calcitonin ([125I]sCT: 74 TBq/mmol) was obtained from Amersham (Buckinghamshire, UK). The monoclonal antibodies (mAbs) used in this study are mouse mAbs to rat LFA-1{alpha}-chain (CD11a), LFA-1ß-chain (CD18), ICAM-1 (CD54), MHC class1, granulocytes and macrophages, pan B-cells, pan T-cells (CD5), helper T cells (CD4), activated T cells, Thy1, killer/suppressor cells (CD8), and Kat1-antigen. All of these mAbs, except mAb anti-Kat1, were purchased from Seikagaku Kogyo (Tokyo, Japan). MAb Kat1 was prepared in our laboratory as described previously (10). Biotinylated antimouse IgG (H+L), IgM, and the Vectastain ABC-AP staining kit were purchased from Vector Laboratories (Burlingame, CA). NR-M2, Konicadol, and Konifix were fom Konica Co. (Tokyo, Japan). Rat antimouse {gamma} PE was from Becton Dickinson (San Jose, CA). Fluorescein isothiocyanate isomer-1 (FITC) was from Dojin Chemical (Kumamoto, Japan). Mouse IgM negative control FITC was from YLEM (Roma, Italy). Actinomycin-D was obtained from Sigma Chemicals.

Bone marrow cultures
Rat bone marrow cells were obtained from the tibiae and femurs of 4- to 6-week-old rats and cultured under the conditions described below, followed by staining with the cytochemical staining kit for TRAP, the autoradiography of CTR, and/or immunostaining with various mAbs.

Whole bone marrow cell cultures for forming osteoclast-like MNCs with high bone-resorbing activity. The bone marrow cells obtained were cultured in 24-well culture plates (106 cells/well) in the presence of 10% heat-treated conditioned medium derived from ROS17/2.8 cells (htROSCM) and 10-8 M 1{alpha},25-(OH)2D3 for 4 days, according to the method of Kukita et al. (12); this preparation yielded approximately 2–5 x 102 bone resorbing MNCs per well when bone marrow cells (106 cells/well) were cultured for 4 days. In the latter part of this study, various concentrations of LFA-1/ICAM-1-blocking peptide with the sequence of the binding site of ICAM-1 against the ligand LFA-1 were added to the cultures. TRAP-positive MNCs with more than three nuclei were counted under a light microscope.

Stromal cell-deprived bone marrow cultures for forming preosteoclast-like cells. Bone marrow cells were deprived of stromal cells by use of a Sephadex G-10 column, as described previously (9). Briefly, bone marrow cells (approximately 1–2 x 108; in 1.5-ml {alpha}MEM containing 15% FCS) were applied to the Sephadex G-10 column (15 ml in 30 ml syringe) equiblirated with {alpha}MEM containing 15% FCS, to remove adherent cells, thus providing nonadherent bone marrow cells (NABMCs; approximately 6–8 x 107cells). These NABMCs did not form any colonies of stromal cells when they were seeded into 35-mm dishes at 106 cells/dish and cultured in {alpha}MEM containing 15% FCS for 8 days. NABMCs (2 x 107; in 10 ml {alpha}MEM containing 15% FCS) were seeded into 100-mm culture dishes, or 106 cells (in 500 µl {alpha}MEM containing 15% FCS) were seeded into 24-well culture plates in the presence of 10% htROSCM and 10-8 M 1{alpha},25(OH)2D3 for 5 days with or without LFA-1/ICAM-1-blocking peptide.

Preparation of rat primary osteoblasts
Rat primary osteoblasts were isolated by sequential digestion from the calvaliae of Sprague-Dawley rats, according to the method of Takahashi et al. (13). The cells were cultured in {alpha}MEM containing 15% FCS in 100-mm culture dishes at 3 x 105 cells/dish. After 4 days of culture, the cells were trypsinized and used as primary osteoblasts.

Double detection by autoradiography using [125I]sCT and immunostaining using mAbs
Cultured cells [MNCs or preosteoclast (POC)-like cells] in 24-well plates were incubated for 1 h at room temperature with 1 µCi/ml (1.77 ng/ml) [125I]sCT. After the removal of the culture medium by aspiration, the cells were incubated with the mAbs described above for 30 min on ice. The cells were rinsed once with ice-chilled {alpha}MEM and fixed with 2% glutaraldehyde and 2% formaldehyde in 0.1 M cacodylate buffer (ph 7.3). After the cells were washed with 0.1 M cacodylate buffer (ph 7.3) 4 times, they were incubated with 3% goat serum in PBS at 4 C overnight to block the nonspecific binding. After allowing for reaction with biotinylated antimouse IgG or IgM antibodies for 30 min, the cells were processed for the avidin-biotin complex procedure using the ABC-AP kit. The endogenous alkaline phosphatase activity was blocked with 1 mm levamisole. After being fixed with 5% glutaraldehyde in PBS at room temperature for 30 min and rinsed with distilled water 4 times, the cells were air-dried. The bottom of each well was cut out from the culture plates, dipped in NR-M2 emulsion, and air-dried, followed by exposure for 10 days at 4 C. Autoradiographs were developed with Konicadol and fixed with Konifix.

Flow cytometry analysis
After 5 days of culture, POC-like cells in 10-cm dishes were removed by 0.02% EDTA in PBS. Cell surface phenotyping was performed by the double staining of direct immunofluorescence using FITC-mAb Kat1, and indirect immunofluorescence using anti-LFA-1{alpha}, LFA-1ß, ICAM-1, and pan B cells as a negative control, and PE-labeled rat antimouse IgG {gamma} chains as the revealing antibodies. The indirect immunofluorescence was performed first. NABMCs (5 x 105) were incubated with 10 µg/ml mAbs for 30 min on ice. After three washes with PBS 2% FCS, the cells were incubated for 30 min on ice with 5 µg/ml of revealing antibodies. Indirect immunofluorescence was done next. After three washes with PBS 2% FCS, the cells were incubated for 30 min on ice with 20 µg/ml FITC-anti-Kat1. Irrelevant mouse FITC-labeled IgM mAb (IgM negative control for mAb anti-Kat1) was included as a control. After three washes with PBS 2% FCS, the cells were treated with actinomycin-D (20 µl/ml). Immunofluorescence was analyzed using a FACScan analyzer (Becton Dickinson).

Design of LFA-1/ICAM-1-blocking peptide
ICAM-1 contains five tandem Ig-like domains (14, 15). Stauton et al. (16) reported that domain 1 of human ICAM-1 contains the primary site of contact for LFA-1, and the mutation of glutamic acid 34 (E34) to alanine (A) in domain 1 completely eliminates LFA-1 binding. The amino acid residue E34 (E35 in the rat) (regarded as essential for binding to LFA-1 in human ICAM-1, mouse ICAM-1, and human ICAM-2) is also conserved in rat ICAM-1 (17). We therefore designed the ICAM-1 peptide fragment composed of five amino acid residues (LGLET) with this important residue (E35) (LFA-1/ICAM-1-blocking peptide) (Fig. 1Go). As a control peptide, we designed the pentapeptide (LGLAT), in which E35 was replaced by A in the above ICAM-1 peptide.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 1. Design of LFA-1/ICAM-1-blocking peptide. Alignment of rat, mouse, and human ICAM-1 amino acid sequences (domain 1). Domain 1 of human ICAM-1 contains the primary site of contact for LFA-1. The amino acid residue E34 (E35 in the rat), regarded as essential in binding to LFA-1 in human ICAM-1, mouse ICAM-1, and human ICAM-2, is also conserved in rat ICAM-1. We designed the ICAM-1 peptide fragment composed of five amino acid residues (LGLET), including E35 (LFA-1/ICAM-1-blocking peptide). The LFA-1/ICAM-1-blocking peptide is boxed. This peptide is thought to block the interaction between LFA-1 and ICAM-1. Each dot represents an amino acid residue that is identical to the residue shown for rat ICAM-1. Gaps (-) have been inserted to achieve maximum homology.

 
Coculture and dentin resorption assay
NABMCs (1 x 106 cells; in 500 µl {alpha}MEM containing 15% FCS) were seeded into 24-well culture plates in the presence of 10% htROSCM and 10-8 M 1{alpha},25(OH)2D3 with the control peptide or the LFA-1/ICAM-1-blocking peptide for 5 days. Then rat primary osteoblasts (103 cells/well) were added to each well. After 4 days of coculture, cells were detached from the culture plates with 0.05% trypsin and 0.02% EDTA in PBS and replated onto dentin slices. Detection and estimation of the resorption pits were performed as described previously (9, 10, 12).

Statistical analysis
All data obtained from bone marrow cultures were analyzed by use of both Student’s t test and post-ANOVA test.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Surface phenotype analysis of mononuclear cells and MNCs formed in culture of osteoclastogenesis using immunocytochemistry
To determine what types of cells are involved in the cultures used for forming preosteoclast-like cells or osteoclast-like MNCs, immunocytochemical staining was performed using commercially available mAbs recognizing different hematopoietic cell lineages. Table 1Go shows the surface phenotyping of the mononuclear cells and that of the MNCs formed in the rat bone marrow cultures. In the cultures for forming preosteoclast-like cells, the mononuclear cells were stained with the mAbs specific to ICAM-1, LFA-1{alpha}, LFA-1ß, and MHC class1. Although these mononuclear cells were stained by the mAb recognizing granulocytes and macrophages, they failed to react with mAbs specific to lymphocytes (pan B, pan T, helper T, and activated T cells). In addition, the mononuclear cells observed in the POC cultures were not stained with mAbs recognizing the cell surface antigens Thy1 and CD8, respectively. A similar surface phenotype was observed in the MNCs, except for the reactivity of anti-LFA-1{alpha} and anti-LFA-1ß mAbs. The MNCs failed to react with these antibodies, whereas the mononuclear cells in the POC cultures for forming preosteoclasts were positive to these antibodies.


View this table:
[in this window]
[in a new window]
  		Table 1. Cell surface phenotypes of mononuclear cells and MNCs in bone marrow cultures

 
Expression of LFA-1 and ICAM-1 on preosteoclasts
Because ICAM-1 is the ligand of LFA-1 and the binding of these two cell surface molecules promotes a wide variety of cellular interactions (14, 15, 18, 19, 20), we further examined whether these cell surface molecules are actually expressed on the preosteoclasts. We used two cell surface markers for identifying preosteoclasts. One is the CTR, the other is the Kat1-antigen. The former marker is widely used as the most reliable marker for identifying cells as osteoclasts, and the latter marker is specifically expressed on cells in the osteoclast-lineage of rats, which is identified by anti Kat1-Ag mAb Kat1. Because almost 100% of Kat1-Ag-expressing mononuclear cells coexpress CTR, Kat1-Ag is thought to be a reliable new marker for identifying cells in the osteoclast lineage (11, 21) (Kukita et al., manuscript in preparation).

Immunohistochemical staining with mAbs and autoradiography were performed simultaneously on rat preosteoclast-like cells to determine whether LFA-1 and ICAM-1 were expressed on preosteoclasts. LFA-1{alpha}, LFA-1ß, and ICAM-1 were expressed on some of the CTR-positive cells (Fig. 2Go). The expression rates of each cell-surface antigen in the CTR-positive cells were as follows: LFA-1{alpha}, 33.5%; LFA-1ß, 24.5%; and ICAM-1, 32.2%.



View larger version (99K):
[in this window]
[in a new window]
  		Figure 2. Detection of LFA-1{alpha}, LFA-1ß, and ICAM-1 on preosteoclast-like cells expressing CTR. Stromal cell-deprived bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3. At 5 days of culture, cells were bound with [125I]sCT, followed by immunostaining with anti-LFA-1{alpha}, ß, and ICAM-1 antibodies. After color development, these cells were processed for autoradiography, as described in Materials and Methods. LFA-1{alpha} (A and B), LFA-1ß (C and D), and ICAM-1 (E and F) were detected on some of the mononuclear cells expressing CTR. A, C, and E are in the light micrographs; and B, D, and F are in the dark micrographs. Double-positive cells are indicated by arrows. The cells in all six panels were photographed at x400 magnification.

 
A three-color analysis was also performed using the FACScan flow cytometry. Cells were initially gated by high forward- and side-scatter, then fractionated to smaller (fraction A) and larger (fraction B) cells (Fig. 3Go). Because the larger cells have autofluorescence, it was difficult to analyze these cells. Kat1-positive cells were detected mainly in the smaller cell fraction (fraction A), which contained 54.1% of the cells in the A and B fractions. To elucidate the phenotype of the Kat1-positive cells, fraction A was further analyzed. Because mononuclear cells expressing CTR correspond with those expressing Kat1-antigen, we examined whether LFA-1{alpha}, ß, or ICAM-1 were expressed on preosteoclast-like cells expressing Kat1-antigen. We performed a flow cytometric double-staining analysis of the surface antigens of the preosteoclast-like cells using mAb Kat1 as the marker for identifying preosteoclasts and anti-LFA-1{alpha}, ß, or anti-ICAM-1 mAbs. As shown in Fig. 4Go, double-positive cell populations were detected. The percentages of cells falling within the dot plot quadrants is shown. In the Kat1-positive cells (approximately 14.1–20% of all cells analyzed), the percentages of cells expressing LFA-1{alpha}, LFA-1ß, or ICAM-1 were 23.4%, 27.5%, and 27.2%, respectively. These data demonstrate that preosteoclast-like cells express LFA-1{alpha}, LFA-1ß, and ICAM-1. These results also suggest the presence of these subpopulations in preosteoclasts.



View larger version (48K):
[in this window]
[in a new window]
  		Figure 3. Expression of Kat1 in preosteoclast-like cells. Stromal cell-deprived bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3. At 5 days of culture, after removal from culture dishes, cells were stained with antibodies to Kat1-FITC and actinomycin D, then analyzed using a FACScan analyzer. The dead cells of the populations were gated out by actinomycin-D staining and gated by high forward- and side-scatter, then separated to smaller (fraction A: 50.4%) and larger (fraction B: 42.7%) cells. Kat1-positive cells were detected mainly in the smaller cell fraction (fraction A).

 


View larger version (43K):
[in this window]
[in a new window]
  		Figure 4. Flowcytometric demonstration of the expression of LFA-1{alpha}, LFA-1ß, and ICAM-1 on preosteoclast-like cells. Stromal cell-deprived bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3. At 5 days of culture, after removal from culture dishes, cells were stained with PE-labeled anti-ICAM-1 antibody (ICAM-1-PE), anti-LFA-1{alpha} antibody (LFA-1{alpha}-PE), anti-LFA-1ß antibody (LFA-1ß-PE), and FITC-labeled mAb Kat1, as described in Materials and Methods. Cells were analyzed using an FACScan analyzer; populations were gated out for dead cells by actinomycin-D staining and gated for subpopulations including Kat1-Ag-positive cells. The subpopulations were then analyzed for the surface expression of ICAM-1, LFA-1{alpha}, and LFA-1ß. Irrelevant mouse IgM served as control for anti-Kat1-Ag mAb,; and anti-pan B mAb served as control for mAb anti-LFA-1{alpha}, anti-LFA-1ß, and anti-ICAM-1. The percentages of cells with a particular cell surface expression phenotype are indicated inside the appropriate quadrant. LFA-1{alpha}, LFA-1ß, and ICAM-1 were expressed on 23.4%, 27.5%, and 27.2% of the Kat1-positive cells, respectively. These data show the typical experiment from three independent experiments.

 
Failure to express LFA-1 in osteoclast-like MNCs
The expression of LFA-1 and ICAM-1 in osteoclast-like MNCs was analyzed by double-detection methods using immunocytochemistry and autoradiography. Osteoclast-like MNCs expressing CTR failed to express LFA-1{alpha} and LFA-1ß (Fig. 5Go), whereas they expressed ICAM-1 frequently (89.3%). Similarly, the stromal cells expressed ICAM-1 but did not express a detectable level of LFA-1.



View larger version (97K):
[in this window]
[in a new window]
  		Figure 5. Detection of ICAM-1 but not LFA-1{alpha}, ß, on osteoclast-like MNCs expressing CTR. Bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3. At 4 days of culture, cells were bound with 125I-labeled salmon calcitonin, followed by immunostaining with anti-LFA-1{alpha}, ß, or ICAM-1 antibody. These cells were processed for autoradiography, as described in Materials and Methods. The majority of CTR-positive osteoclast-like MNCs expressed ICAM-1 (E and F). Double-positive cells are indicated by arrows. LFA-1{alpha} (A and B) and LFA-1ß (C and D) were not expressed on these cells. A, C, and E are light micrographs; and B, D, and F are dark micrographs. The cells in all six panels were photographed at x400 magnification.

 
LFA-1/ICAM-1-blocking peptide inhibits osteoclast-like cell formation
To investigate the involvement of the interaction between ICAM-1 and LFA-1 in osteoclastogenesis, we designed a blocking peptide that interupts LFA-1/ICAM-1 binding. It has been reported that amino acid residue E35 of ICAM-1 Ig-like domain 1 is essential for binding to LFA-1. We therefore prepared a peptide composed of five amino acid residues (LGLET: LFA-1/ICAM-1-blocking peptide) with a sequence similar to those around residue E35, as shown in Fig. 1Go. As the control peptide, we also prepared a pentapeptide LGLAT in which essential E35 was displaced by A. We examined the effect of the LFA-1/ICAM-1-blocking peptide on the osteoclast formation. Figure 6Go shows the dosage effect of the LFA-1/ICAM-1-blocking peptide on the formation of TRAP-positive MNCs. A significant suppressive effect of the LFA-1/ICAM-1-blocking peptide was observed in this culture system of osteoclastogenesis, whereas the control peptide did not show any effect on osteoclastogenesis. The maximal suppressive effect was obtained at a concentration of 100 ng/ml. We have also confirmed, as an alternative control protein, that BSA had no effect on MNC formation (data not shown). We also examined whether this inhibitory effect of the LFA-1/ICAM-1-blocking peptide on MNC formation was reversible. As shown in Fig. 7Go, a pulse-treatment of the culture with the blocking peptide delayed the rate of the formation of MNCs; however, a complete recovery was obtained at 5 days of the culture. This result shows that the inhibitory effect of the blocking peptide is reversible and that it is not cytotoxic. These results suggest an involvement of the LFA-1/ICAM-1 interaction in the formation of osteoclast-like MNCs.



View larger version (86K):
[in this window]
[in a new window]
  		Figure 6. Dosage effect of the LFA-1/ICAM-1-blocking peptide on the formation of osteoclast-like MNCs. Bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3, as described in Materials and Methods, with various concentrations of LFA-1/ICAM-1-blocking peptide or control peptide. Wells were stained after 4 full days of culture, and TRAP-positive cells containing more than three nuclei/cell were counted. The LFA-1/ICAM-1-blocking peptide decreased the formation of osteoclast-like cells, with the peak suppressive effect at the concentration of 100 ng/ml. Each bar represents the mean ± SEM of quadruplicate cultures in 24-well culture plates. Data were analyzed by Student’s t test. *, P < 0.05. **, P < 0.01. Data represent a typical experiment from three independent cultures.

 


View larger version (17K):
[in this window]
[in a new window]
  		Figure 7. Complete recovery of MNC formation after pulse-treatment with LFA-1/ICAM-1-blocking peptide. Bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3, as described in Materials and Methods. Cells were pulse-treated with 100 ng/ml LFA-1/ICAM-1-blocking peptide or with control peptide from day 1 to day 2 of the culture. After removal of the peptide at 2 days of the culture, cells were continued to culture by 5 days. The number of TRAP-positive MNCs were counted on each day of the culture. Complete recovery of MNC formation from the inhibition provoked by the LFA-1/ICAM-1 peptide was obtained on 5 days of the culture. Open squares, Cells treated with LFA-1/ICAM-1-blocking peptide; closed diamonds, cells treated with the control peptide.

 
LFA-1/ICAM-1-blocking peptide inhibits both the early stage and late stage of osteoclast-like MNC formation
We examined whether the LFA-1/ICAM-1-blocking peptide affects the early stage or the late stage of the osteoclast-like MNC formation. As shown in Fig. 8Go, with treatment of whole bone marrow cells with the LFA-1/ICAM-1-blocking peptide (100 ng/ml) for 4 days, the first 2 days, or last 2 days of the culture, the formation of osteoclast-like MNC was inhibited by 61.2%, 25.3%, and 31.5%, respectively. The peptide treatment throughout the culture was the most effective. These data show that the LFA-1/ICAM-1 peptide is effective at both the early stage and the late stage of the culture in the presence of the stromal cells. We investigated whether the LFA-1/ICAM-1-blocking peptide affected the formation of preosteoclast-like cells in a bone marrow culture system deprived of stromal cells. We found that the formation of TRAP-positive preosteoclast-like mononuclear cells was not affected by the presence of any concentrations of the LFA-1/ICAM-1-blocking peptide, as shown in Fig. 9Go. We further confirmed that these preosteoclast-like mononuclear cells treated with LFA1/ICAM1-blocking peptide were not altered their ability to form bone-resorbing cells. These preosteoclast-like cells, treated with LFA-1/ICAM-1-blocking peptide, were cocultured with primary osteoblasts to form MNCs, followed by replating onto dentin slices. Fig. 10Go shows resorption lacunae formed by these MNCs. Table 2Go shows a quantitative datum of the dentin resorption using dentin slices with exactly equal diameter. Ability of the formed MNCs to resorb dentin was not changed by treatment of the culture to form preosteoclast-like cells with the LFA1/ICAM1-blocking peptide before coculturing with osteoblasts. In both groups treated with the control peptide or with the blocking peptide, the percentage of (number of pits)/(number of TRAP-positive MNCs) was approximately 65%. These results suggest that the interaction between LFA-1 and ICAM-1 is not essential for preosteoclast formation from bone marrow progenitor cells in the absence of stromal cells.



View larger version (53K):
[in this window]
[in a new window]
  		Figure 8. Effect of the temporal addition of LFA-1/ICAM-1-blocking peptide on the formation of TRAP-positive MNCs. Bone marrow cells were cultured in the presence of htROSCM and 1{alpha},25(OH)2D3, as described in Materials and Methods. Control cells were cultured with 100 ng/ml of control peptide. The culture groups (three experiments, consisting of four wells/group) included the controls: 100 ng/ml LFA-1/ICAM-1-blocking peptide at all times; during the period when preosteoclast proliferation occurs (days 1–2); and during the period when preosteoclast fusion occurs (days 3–4). In all groups, medium was changed at 2 days of culture. Cells were stained for TRAP at 4 days of culture, and the number of TRAP-positive MNCs was counted. The treatment of cultures for days 1–2, for days 3–4, and for all days decreased the osteoclast-like cells by 25.3%, 31.5%, and 61.2% (compared with the control values), respectively. Each bar represents the mean ± SEM of quadruplicate cultures in 24-well culture plates. Data were analyzed by Student’s t test. *, P < 0.05; **, P < 0.01. Data represent a typical experiment for four independent experiments.

 


View larger version (83K):
[in this window]
[in a new window]
  		Figure 9. LFA-1/ICAM-1-blocking peptide does not affect the formation of preosteoclast-like mononuclear cells in stromal cell-deprived marrow cultures. Stromal cell-deprived bone marrow cells were cultured in the presence of various concentrations of LFA-1/ICAM-1-blocking peptide or control peptide (100 ng/ml) for 5 days. Cells were fixed and stained for TRAP. The number of TRAP-positive mononuclear cells was counted in each well. The LFA-1/ICAM-1-blocking peptide did not affect the formation of preosteoclast-like cells. Each bar represents the mean ± SEM of quadruplicate cultures in 24-well culture plates. Data were analyzed by Student’s t test. *, P < 0.05; **, P < 0.01. Data represent a typical experiment from three independent cultures.

 


View larger version (145K):
[in this window]
[in a new window]
  		Figure 10. Demonstration of the resorption pits formed by the osteoclast-like MNCs derived from preosteoclast-like cells treated with the LFA-1/ICAM-1-blocking peptide. Stromal cell-deprived bone marrow cells were cultured in the presence of 100 ng/ml LFA-1/ICAM-1-blocking peptide or control peptide for 5 days. Then rat primary osteoblasts were added to each well. After 4 days of coculture, cells were detached from the culture plates with 0.05% trypsin and 0.02% EDTA in PBS, followed by replating onto human dentin slices, and were cultured for 3 days. These dentin slices were processed for toluidine blue staining, as described in Materials and Methods. Typical resorption lacunae were observed on dentin slices. The resorption lacunae was photographed at x400 magnification.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Pretreatment of cultures for forming preosteoclast-like cells with LFA-1/ICAM-1-blocking peptide does not affect function of MNCs formed after coculture with osteoblasts

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
We studied which lineage existed in an osteoclast-forming culture system, by use of lineage-specific antibodies. Both preosteoclast-like mononuclear cells and osteoclast-like MNCs expressed a granulocyte and macrophage lineage-specific antigen but did not express lymphocyte lineage-specific antigens, except that preosteoclast-like cells expressed LFA-1. These findings are in keeping with previous reports that osteoclasts originate in granulocyte-macrophage-colony forming unit (2, 22, 23, 24, 25). We also observed some cell surface antigens, and we found that a subpopulation expressing LFA-1 and/or ICAM-1 molecules existed in preosteoclast-like mononuclear cells. In the osteoclast-like MNCs, the expression of LFA-1 was not detected, although these cells showed an intense expression of ICAM-1. ICAM-1 is the known ligand of the integrin LFA-1, and the binding of these cell surface molecules to each other promotes a wide variety of cellular interactions (14, 15, 18, 19, 20). ICAM-1 is expressed on the surface of a variety of hematopoietic cells and also nonhematopoietic cells, such as vascular endothelial cells, thymic epithelial cells, tracheal epithelial cells, and fibroblasts (26). The expression of LFA-1, in contrast, has been thought to be restricted to leukocytes (27). To determine whether LFA-1 antigen-positive lymphocytes are involved in these cultures or preosteoclasts themselves express LFA-1, we used a double-detection method on preosteoclast-like mononuclear cells by using anti-LFA-1{alpha}, anti-LFA-1ß, or anti-ICAM-1 mAb and anti-Kat1 mAb. Kat1 antigen is a reliable immunological marker of osteoclast-lineage (osteoclast-like MNCs and their mononuclear precursors) and is recognized by mAb Kat1 (10, 11). We recently successfully used this mAb for the detection of cells in osteoclast-lineage (21, Kukita manuscript in preparation). The present results demonstrated that approximately 30% of the preosteoclasts expressed LFA-1. This flowcytometric analysis also revealed that approximately 30% of the preosteoclasts expressed ICAM-1. These data show that a subpopulation exists expressing {alpha}Lß2 integrin (LFA-1) and/or ICAM-1 in the cell population of the preosteoclasts. The level of ICAM-1 expression did not show a significant change, whereas that of LFA-1 fell to an undetectable level during the multinucleation step. These findings are highly suggestive of the existence of a subpopulation of preosteoclasts in which LFA-1 is transiently expressed only in the stage of preosteoclasts, the mononuclear precursors of osteoclasts. Concerning the subpopulation expressing LFA-1 and ICAM-1, two hypotheses are taken into consideration, as shown in Fig. 11Go. The first hypothesis is that the LFA-1-positive subpopulation does not overlap with the ICAM-1-positive subpopulation. The other hypothesis is that a unique LFA-1 and ICAM-1 double-positive subpopulation exists in the cell population of preosteoclasts. Because the LFA-1/ICAM-1-blocking peptide inhibited the formation of osteoclast-like MNCs by approximately 60% when the blocking peptide was added two times to the culture for forming MNCs (Fig. 8Go), the former hypothesis may be more likely. A four-color FACS analysis and cell sorting are required for elucidating the entity of these subpopulations observed in preosteoclasts. Such a flowcytometric analysis is underway in our laboratories. Because adhesion molecules are thought to be involved in the interactions between osteoclast progenitors and stromal cells or those among preosteoclasts themselves during osteoclastogenesis, the expression of LFA-1 and ICAM-1 on cells in osteoclast-lineage prompted us to perform experiments regarding the actual participation of these two molecules in osteoclastogenesis. The LFA-1/ICAM-1-blocking peptide was shown to have an inhibitory effect on osteoclast-like MNCs formation, in a dose-dependent manner. This inhibition was observed when the LFA-1/ICAM-1-blocking peptide was added for the first 2 days, as well as for the last 2 days of the culture. The NABMC culture system containing no stromal cells was unaffected by the LFA-1/ICAM-1-blocking peptide. With these findings, we surmise not only that the LFA-1/ICAM-1-dependent pathway is involved in fusion among preosteoclasts but also that LFA-1 is already expressed on osteoclast progenitors, and this pathway plays a role in interactions between stromal cells and osteoclast progenitors (Fig. 11Go). There have been no detailed reports referring to the expressional stage of LFA-1 in osteoclastogenesis. Our present study obtained direct evidence concerning the expression of LFA-1 and ICAM-1 on preosteoclasts. The results in our study, using the LFA-1/ICAM-1-blocking peptide, are almost in agreement with the experiment using mAb to LFA-1 or ICAM-1, reported by Kurachi et al. (6). However, in their study, anti-LFA-1 and anti-ICAM-1 antibodies almost completely inhibited formation of cells having TRAP activity at high doses. In contrast, our result revealed that the inhibitory effect provoked by LFA-1/ICAM-1-blocking peptide was less than 61%. We think that LFA-1/ICAM-1 interaction could provide one of the mechanisms concerning the cell-cell interactions involved in the osteoclastogenesis. Feuerbach et al. (3, 4) reported that expression of VCAM-1 in stromal cells is required for osteoclastogenesis. Mbalaviele et al. (7) showed the participation of E-cadherin in the fusion process of preosteoclasts. Involvement of other adhesion molecules in osteoclast differentiation remains to be determined.



View larger version (44K):
[in this window]
[in a new window]
  		Figure 11. Expression and roles of LFA-1 and ICAM-1 in osteoclastogenesis: two hypotheses. The first hypothesis is that the LFA-1-positive subpopulation does not overlap with the ICAM-1-positive subpopulation, as shown in the left flow. Cells expressing LFA-1 bind to ICAM-1 on stromal cells and fuse with cells expressing ICAM-1, followed by the differentiation into multinucleated osteoclast-like cells. The other hypothesis is that a double-positive subpopulation expressing both LFA-1 and ICAM-1 exists in the cell population of preosteoclasts, as shown in the right flow. These double-positive cells bind to stromal cells and fuse to each other to form multinucleated osteoclast-like cells. The expression of LFA-1 is selectively diminished, whereas that of ICAM-1 is retained during the step of multinucleation.

 
In view of the clinical aspects, it is well known that ICAM-1 has a close relationship with inflammation. Therapeutic approaches have been made to induce antiinflammatory effects by the blocking of the LFA-1/ICAM-1-dependent pathway, using a neutralizing antibody, in experimental animal models of various human diseases such as pneumonia (28), nephritis (29), arthritis (30), and asthma (31), among others. In the United States, clinical trials, using antihuman ICAM-1 Ab, have been performed; and several reports on its clinical effects have been published (32). Ten patients with rheumatoid arthritis were treated with a total of 140 or 280 mg of murine anti-CD54 mAb iv over 5 days. A rapid clinical improvement was noted. In rheumatoid arthritis, the blockage of the interaction between LFA-1 and ICAM-1 may inhibit not only inflammation in the joints but also bone resorption, by suppressing the formation of osteoclasts. The LFA-1/ICAM-1-blocking peptide designed in this study is expected to provoke an inhibitory effect on the LFA-1/ICAM-1 pathway; thus, it is anticipated that an inhibitory effect of bone resorption may occur through the suppression of osteoclastogenesis. The LFA-1/ICAM-1-blocking peptide may be useful in the clinical treatment of metabolic bone diseases such as rheumatoid arthritis, osteoporosis, and Paget’s disease.


   Acknowledgments
 
The authors thank Drs. Yoichi Sugioka, Takao Hotokebuchi, Seiya Jingushi, and Toshihide Shuto (of Kyushu University, Faculty of Medicine, Department of Orthopedics) for encouragement. We also thank Dr. Seiji Nakamura, Shinichiro Yada, Takahiko Nakamura, Kouji Tamada, and Hisakata Yamada (of Kyushu University, Faculty of Dentistry and Medicine) for helpful discussion.

Received January 14, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Burger EH, Nijweide PJ 1991 The osteoclast. In: Brian KH (ed) Bone. CRC Press, Boca, Raton Vol 2:31–59
  2. Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocr Rev 13:66–80[Abstract/Free Full Text]
  3. Feuerbach D, Feyen JHM 1996 Expression of the cell-adhesion molecule VCAM-1 by stromal cells is necessary for osteoclastogenesis. FEBS Lett 402:21–24
  4. Feuerbach D, Loetscher E, Buerki K, Sampath TK, Feyen JHM 1997 Establishment and characterization of conditionally immortalized stromal cell lines from a temperature-sensitive T-Ag transgenic mouse. J Bone Miner Res 12:179–190[CrossRef][Medline]
  5. Duong LT, Tanaka H, Rodan GA VCAM-1 involvement in osteoblast-osteoclast interaction during osteoclast differentiation. Program of the 16th Annual Meeting of The American Society for Bone and Mineral Research, Kansas City, MO, 1994, p 131 (Abstract)
  6. Kurachi K, Morita I, Murota S 1993 Involvement of adhesion molecules LFA-1 and ICAM-1 in osteoclast development. Biochim Biophys Acta 1178:259–266[Medline]
  7. Mbalaviele G, Chen H, Boyce BF, Mundy GR, Yoneda T 1995 The role of cadherin in the generation of multinucleated osteoclasts from mononuclear precursors in murine marrow. J Clin Invest 95:2757–2765
  8. Kumegawa M 1987 The bone - the first cinematographic record of osteoclast formation. Program of the 9th Annual Meeting of The American Society for Bone and Mineral Research, Indianapolis IN, 1987, p 269 (Abstract)
  9. Kukita A, Kukita T, Shin JH, Kohashi O 1993 Induction of mononuclear precursor cells with osteoclastic phenotypes in a rat bone marrow culture system depleted of stromal cells. Biochem Biophys Res Commun 196:1383–1389[CrossRef][Medline]
  10. Kukita T, Kukita A, Nagata K, Maeda H, Kurisu K, Watanabe T, Iijima T 1994 A novel cell surface antigen expressed on rat osteoclasts regulating the function of the calcitonin receptor. J Immunol 153:5265–5273[Abstract]
  11. Kukita T, Kukita A, Xu LX, Maeda H, Iijima T Successful detection of active osteoclasts in situ by systemic administration of an osteoclast-specific monoclonal antibody. Calcif Tissue Int, in press
  12. Kukita A, Kukita T, Hata K, Kurisu K, Kohashi O 1993 Heat-treated osteoblastic cell (ROS17/2.8)-conditioned medium induces the formation of osteoclast-like cells. Bone Miner 23:113–127[Medline]
  13. Takahashi N, Akatsu T, Udagawa N, Sasaki T, Yamaguchi A, Moseley JM, Martin TJ, Suda T 1988 Osteoblastic cells are involved in osteoclast formation. Endocrinology 123:2600–2602[Abstract/Free Full Text]
  14. Dustin ML, Springer TA 1991 Role of lymphocyte adhesion receptors in transient interactions and cell locomotion. Annu Rev Immunol 9:27–66[CrossRef][Medline]
  15. Springer TA 1990 Adhesion receptors of the immune system. Nature 346:425–434[CrossRef][Medline]
  16. Stauton DE, Dustin ML, Erickson HP, Springer TA 1990 The arrangement of the immunoglobulin-like domains of ICAM-1 and the binding sites for LFA-1 and Rhinovirus. Cell 61:243–254[CrossRef][Medline]
  17. Kita Y, Takashi T, Iigo Y, Tamatani T, Miyasaka M, Horiuchi T 1992 Sequence and expression of rat ICAM-1. Biochim Biophys Acta 1131:108–110[Medline]
  18. Inaba K, Steinman RM 1984 Resting and sensitized T lymphocytes exhibit distinct stimulatory (antigen-presenting cell) requirements for growth and lymphokine release. J Exp Med 160:1717–1735[Abstract/Free Full Text]
  19. Osborn L 1990 Leukocyte adhesion to endothelium in inflammation. Cell 62:3–6[CrossRef][Medline]
  20. Carlow DA, Oers NSC, Teh SJ, Teh S 1992 Deletion of antigen-specific immature thymocytes by dendritic cells requires LFA-1/ICAM-1 interactions. J Immunol 148:1595–1603[Abstract]
  21. Xu LX, Kukita T, Kukita A, Otsuka T, Niho Y, Iijima T 1995 Interleukin-10 selectively inhivits osteoclastogenesis by inhibiting differentiation of osteoclast progenitors into preosteoclast-like cells in rat bone marrow culture system. J Cell Physiol 165:624–629[CrossRef][Medline]
  22. Kurihara N, Suda T, Miura T, Nakauchi H, Kodama H, Hiura K, Hakeda Y, Kumegawa M 1989 Generation of osteoclasts from isolated hematopoietic progenitor cells. Blood 74:1295–1302[Abstract/Free Full Text]
  23. Kurihara N, Chenu C, Miller M, Civin C, Roodman GD 1990 Identification of committed mononuclear precursors for osteoclast-like cells formed in long term human marrow cultures. Endocrinology 126:2733–2741[Abstract/Free Full Text]
  24. Kurihara N, Gluck S, Roodman GD 1990 Sequential expression of phenotype markers for osteoclasts during differentiation of precursors for multinucleated cells formed in long term human marrow cultures. Endocrinology 127:3215–3221[Abstract/Free Full Text]
  25. Udagawa N, Takahashi N, Akatsu T, Tanaka H, Sasaki T, Nishihara T, Koga T, Martin TJ, Suda T 1990 Origin of osteoclasts: mature monocytes and macrophages are capable of differentiating into osteoclasts under a suitable microenvironment prepared by bone marrow-derived stromal cells. Proc Natl Acad Sci USA 87:7260–7264[Abstract/Free Full Text]
  26. Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer TA 1986 Induction by IL 1 and interferon-{gamma}: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). J Immunol 137:245–254[Abstract]
  27. Kishimoto TK, Larson RS, Corbi AL, Dustin ML, Stauton DE, Springer TA 1989 The leukocyte integrins: LFA-1, Mac-1, and p150,95. Adv Immunol 46:149–182[Medline]
  28. Mulligan MS, Wilson GP, Todd RF, Smith CW, Anderson DC, Varani J, Issekutz TB, Miyasaka M, Tamatani T, Rusche JR, Vaporciyan AA, Ward PA 1993 Role of ß1, ß2 integrins and ICAM-1 in lung injury after deposition of IgG and IgA immune complexes. J Immunol 150:2407–2417[Abstract]
  29. Kawasaki K, Yaoita E, Yamamoto T, Tamatani T, Miyasaka M, Kihara I 1993 Antibodies against intercellular adhesion molecule-1 and lymphocyte function-associated antigen-1 prevent glomerular injury in rat experimental crescentic glomerulonephritis. J Immunol 150:1074–1083[Abstract]
  30. Iigo Y, Takashi T, Tamatani T, Miyasaka M, Higashida T, Yagita H, Okumura K, Tsukada W 1991 ICAM-1-dependent pathway is critically involved in the pathogenesis of adjuvant arthritis in rats. J Immunol 147:4167–4171[Abstract]
  31. Wegner CD, Gundel RH, Reilly P, Haynes N, Letts G, Rothlein R 1990 Intercellular adhesion molecule-1 (ICAM-1) in the pathogenesis of asthma. Science 247:456–459[Abstract/Free Full Text]
  32. Kavanaugh AF, Davis LS, Jain RI, Nichols LA, Norris SH, Lipsky PE 1996 A phase I/II open label study of the safety and efficacy of an anti-ICAM-1 (intercellular adhesion molecule-1; CD54) monoclonal antibody in early rheumatoid arthritis. J Rheumatol 23:1338–1344[Medline]



This article has been cited by other articles:


Home page
 JBJSHome page
S. C. Ramage, N. H. Urban, W. A. Jiranek, A. Maiti, and M. J. Beckman
Expression of RANKL in Osteolytic Membranes: Association with Fibroblastic Cell Markers
J. Bone Joint Surg. Am., April 1, 2007; 89(4): 841 - 848.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
T. Iwamoto, S. Fukumoto, K. Kanaoka, E. Sakai, M. Shibata, E. Fukumoto, J.-i. Inokuchi, K. Takamiya, K. Furukawa, K. Furukawa, et al.
Lactosylceramide Is Essential for the Osteoclastogenesis Mediated by Macrophage-Colony-stimulating Factor and Receptor Activator of Nuclear Factor-kappa B Ligand
J. Biol. Chem., November 30, 2001; 276(49): 46031 - 46038.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Harada, H.
Right arrow Articles by Iijima, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harada, H.
Right arrow Articles by Iijima, T.
Right arrowPubmed/NCBI databases
*Substance via MeSH

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals