| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
B Ligand and Their Differential Expression in Bone and Thymus1
Department of Pathology and Immunology (T.I., M.U., K.H.), Aging and Developmental Science, Graduate School, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo 113-8519, Japan; and Department of Bacterial and Blood Products (M.K.), National Institute of Infections Diseases, Shinjuku-ku, Tokyo 162, Japan
Address all correspondence and requests for reprints to: Tohru Ikeda, Department of Pathology and Immunology, Aging and Developmental Science, Graduate School, Tokyo Medical and Dental University, 1–5-45 Yushima, Bunkyo-ku, Tokyo 113-8519, Japan. E-mail: toru.pth2{at}med.tmd.ac.jp
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B ligand [RANKL; also
known as tumor necrosis factor-related activation-induced cytokine,
osteoprotegerin ligand, and osteoclast differentiation factor] is
known to bind with the receptor activator of NF-B (RANK) and act not
only as a key factor for osteoclastogenesis but also as a regulator of
lymphocyte development. In this study, we found two additional isoforms
of RANKL. RANKL 2 has a shorter intracellular domain than the original
RANKL (RANKL 1), and RANKL 3 lacks a transmembrane domain and was
thought to act as a soluble form. In the bone marrow stromal cell line
ST2 and preosteoblastic cell line MC3T3-E1, all three RANKL isoforms
were detected, but the expression of RANKL 2 was preferentially
suppressed by treatment with 1
,25-dihydroxyvitamin D3
and dexamethasone. In young adult thymus,
CD4-CD8- double-negative cells were positive
for all three isoforms, CD4+CD8+
double-positive cells were positive for RANKL 1 and RANKL 3 but
negative for RANKL 2, and CD4+CD8- and
CD4-CD8+ single-positive cells were positive
for all three isoforms. Immunofluorescence analyses of NIH3T3 cells
transfected with each RANKL isoform indicated that the three RANKL
isoforms were translated, and RANKL 2 protein predominantly stayed in
the endoplasmic reticulum and Golgi networks. These results indicate
that there are three kinds of RANKL-RANK pathways. The presence of
multiple RANKL-RANK pathways suggests a more complicated RANKL-RANK
system for osteoclastogenesis or T cell differentiation than previously
thought. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
B ligand
(RANKL) (2), osteoprotegerin ligand (OPGL)
(3), and osteoclast differentiation factor (ODF)
(4). TRANCE and RANKL were isolated in laboratories of
immunology as a factor playing a role in the survival and activation of
dendritic cells (1, 2, 5) or T cells (6).
OPGL and ODF were isolated from a myelomonocytic cell line and bone
marrow stromal cell line, respectively, as a factor of
osteoclastogenesis. RANKL (TRANCE, OPGL, ODF) is known to bind with two
different receptors. One is the receptor activator of NF-B (RANK),
which was isolated from dendritic cells (2). The other is
osteoprotegerin (OPG), a secreted TNF receptor-related protein, which
was isolated as a protein inhibiting bone resorption (7, 8).
Osteoclastogenesis can be efficiently reproduced by the coculture of
bone marrow-derived stromal cells and bone marrow-derived macrophages
or spleen cells in the presence of 1,25-dihydroxyvitamin
D3
(1,25(OH)2D3) and
dexamethasone (Dex). Without bone marrow-derived stromal cells, no
osteoclasts formed from bone marrow-derived macrophages or spleen
cells, even in the presence of
1,25(OH)2D3 and Dex, and
the stromal cells were proved to be essential for osteoclastogenesis
(9, 10, 11). However, the molecular mechanism of in
vitro osteoclastogenesis had been unknown before the discovery of
RANKL. Bone marrow-derived macrophages or spleen cells differentiated
into osteoclasts without stromal cells when the soluble form of
recombinant RANKL protein was added to the culture medium, and RANKL
was strongly suggested to be a key factor in osteoclastogenesis
(3, 4, 12, 13, 14).
The physiological function of RANKL was confirmed in mice with a
disrupted opgl gene (15). The
opgl-deficient mice showed severe osteopetrosis and a defect
in tooth eruption, and they completely lacked osteoclasts as a result
of an inability of osteoclastogenesis. Interestingly, the mice also
exhibited defects in early differentiation of T and B lymphocytes and
lacked all lymph nodes. A recent study by the same group indicated that
T cells were associated with osteoclastogenesis in inflammatory bone
resorption (16). The data directly connected immunology
and bone cell biology. Recently, the presence of another pathway of
osteoclastogenesis, via TNF-, was reported (17).
TNF--induced osteoclasts were thought to play an important role in
bone resorption in inflammatory bone diseases, together with
interleukin-1. The report clearly indicated the presence of two
pathways of osteoclastogenesis; the RANKL-pathway and the
TNF--pathway.
Here we show that two more RANKL isoforms are present. The expression
of three RANKL isoforms is differentially regulated by steroids in
bone-derived cell lines. Furthermore, the three RANKL isoforms are
differentially expressed in T cell subsets during the differentiation
in thymus. These findings indicate that ligand-receptor systems in
osteoclastogenesis are composed of three kinds of RANKL-RANK systems
and TNF--TNF-receptor systems. Furthermore, the data raise the new
hypothetical concept that the regulation of each RANKL isoform is
important for control of bone metabolism and T cell
differentiation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
minimum essential medium (-MEM) supplemented with
10% FCS. Some of the cultures were treated with 1 x
10-8 M
1,25(OH)2D3 and 1 x
10-7 M Dex for
5 days (9). Total RNA of these cells was extracted by
guanidium thiocyanate/cesium chloride centrifugation
(20).
Preparation of thymocytes
T cell subsets of double-negative
CD4-CD8-, double-positive
CD4+CD8+, and
single-positive CD4+CD8-
and CD4-CD8+ were prepared
from the thymus of 4-week-old male C57BL/6 mice. Cell suspensions of
whole thymocytes were first treated with either anti-CD4 or anti-CD8
antibody plus complement and then washed with medium. The cell
suspensions were then stained with fluorescent
isothiocyanate-conjugated anti-CD4 antibody and phycoerythrin
(PE)-conjugated anti-CD8 antibody, and T cell subsets of
CD4-CD8-,
CD4+CD8-, and
CD4-CD8+ were sorted by
flow cytometer (FACS Vantage, Becton Dickinson and Co.,
Mountain View, CA). Cell suspensions of whole thymocytes were stained
with PE-conjugated anti-CD4 antibody and fluorescent
isothiocyanate-conjugated anti-CD8 antibody, and the fraction
CD4+CD8+ was then sorted by
flow cytometer.
Northern hybridization
The thymus, lung, testis, and calvaria were dissected from male
C57BL/6 mice at 6 weeks of age, and homogenized. Total RNA was
extracted in the same manner as described above. Twenty micrograms of
total RNA was loaded in each lane of a 1.2% formaldehyde agarose gel,
transferred onto Hybond-N membrane (Amersham Pharmacia Biotech, Buckinghamshire, UK), fixed with UV, and hybridized at
65 C for 3 h in Rapid hybridization buffer (Amersham Pharmacia Biotech). RANKL 3 complementary DNA (cDNA) fragments
were used as probe and labeled with
[32P]deoxycycidine triphosphate (Amersham Pharmacia Biotech). The membrane was washed at 60 C for 20 min
with 2 x SSC/0.1% SDS, then 20 min twice with 0.2 x
SSC/0.1% SDS at 60 C, and exposed to a BAS imaging plate (Fuji Photo Film Co., Ltd., Kanagawa, Japan). The image was
analyzed with a BAS 2000 imaging analyzer (Fuji Photo Film Co., Ltd.).
cDNA synthesis and RT-PCR
Poly (A)+ RNA was isolated from total RNA
of the organs and ST2 and MC3T3-E1 cells using Oligotex dT-30 super
(Daiichi Kagakuyakuhin Co., Tokyo, Japan), and the
cDNA was synthesized with ReverTra Ace reverse transcriptase (TOYOBO
Co., Osaka, Japan). Total RNA was extracted from thymocytes
using ISOGEN (Wako Pure Chemical Industries, Ltd., Osaka,
Japan). The cDNA was synthesized with ReverTra Ace reverse
transcriptase (TOYOBO Co.) from the total RNA. RT-PCR for cloning of
RANKL isoforms was carried out with 40 cycles of denaturation at 98 C
for 20 sec and annealing and extension at 68 C for 2 min, using LA
Taq polymerase (Takara Co., Shiga, Japan) and the
following primers, which were made, referred to the sequences of mouse
RANKL cDNA (accession numbers AF019048, AB008426, AF053713, and
AF013170): R-5; 5'-ATGCGCCGGGCCAGCCGAGACTACGGC-3' R-3;
5'-TCAGTCTATGTCCTGAACTTTGAAAGCCCC-3'
To detect the expression of each RANKL isoform, RT-PCR was performed with 40 cycles of denaturation at 98 C for 20 sec, annealing at 60 C for 30 sec, and extension at 72 C for 1 min using Taq polymerase (Roche Molecular Biochemicals, Mannheim, Germany) and the following sets of primers, which amplify an 881-nucleotide, 830-nucleotide, and 740-nucleotide product, respectively: RANKL 1: R1-5; 5'-TCCCACACGAGGGTCCGCTG-3' R-3; 5'-TCAGTCTATGTCCTGAACTTTGAAAGCCCC-3' RANKL 2: R2-5; 5'-TGCGCACTCCGGCGTCCCGC-3' R-3; 5'-TCAGTCTATGTCCTGAACTTTGAAAGCCCC-3' RANKL 3: R3-5; 5'-CCGAGACTACGGCGGATCCTAACA-3' R-3; 5'-TCAGTCTATGTCCTGAACTTTGAAAGCCCC-3'
In vitro expression of RANKL cDNA
The protein-coding region of RANKL 1, RANKL 2, and RANKL 3 was
cloned into a selectable mammalian expression vector, pMIKHyg B, which
drives inserted cDNA by the SR promoter (21). Each
construct was transfected into NIH3T3 cells using TransFast
Transfection Reagent (Promega Corp., Madison, WI) and
cultured in -MEM supplemented with 10% FCS. One week after
transfection, stable transfectants of the cells were selected in medium
further supplemented with 250 µg/ml Hygromycin B
(Calbiochem, San Diego, CA). The expression of transfected
cDNA was analyzed by Northern hybridization as described above.
Production of a rabbit polyclonal antiserum
The mouse RANKL 3 peptide (residues 118–316 in RANKL 1 and
89–287 in RANKL 2) was produced using GST Gene Fusion System
(Amersham Pharmacia Biotech). Rabbits were inoculated with
0.25 ml of the peptide (50 mg) emulsified with 0.25 ml complete
Freund’s adjuvant (Difco Lab., Detroit, MI) for four times every four
weeks. Preimmune serum of each animal was taken before the inoculation.
Antisera obtained after the inoculation were confirmed to contain
RANKL-specific antibodies, by enzyme-linked immunosorbent assay and
Western blot analysis using the RANKL 3 peptide.
Immunofluorescence microscopy
Stable transfectants were cultured, at 37 C, in -MEM
containing 10% FCS, on 12-hole heavy Teflon-coated slides (Bokusui
Brown, New York, NY). They were washed twice with PBS containing 0.1
mg/ml CaCl2 and 0.1 mg/ml
MgCl2, for 10 min at room temperature, and then
fixed with cold acetone for 10 min on ice. They were washed three times
with PBS and reacted with the anti-RANKL antiserum (10 µg/ml in PBS)
for 1 h at room temperature. After three more washes with PBS,
they were reacted with TRITC-conjugated swine antirabbit IgG (diluted
100-fold) (DAKO Corp., Glostrup, Denmark) for
1 h at room temperature. The stained samples were washed twice
with PBS and sealed with PBS-glycerin (1:9). Immunofluorescence
photographs were taken on a confocal laser scanning
inverted-microscope, LSM510 (Carl Zeiss, Jena,
Germany).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
Surprisingly, in all these samples, multiple RANKL-like fragments with
different sizes were detected (data not shown). Sequence analyses of
these fragments indicated that there are two new isoforms of RANKL in
addition to the original RANKL (RANKL 1), and the new isoforms were
named RANKL 2 and RANKL 3 (Fig. 1). RANKL 2 had the sequence encoding
the predicted transmembrane domain (Fig. 1, shaded), but the
sequence of the more upstream region was largely deleted (Fig. 1). On
the other hand, in the shortest isoform, RANKL 3, the region from
A29 to G225 in the RANKL 1
nucleotide sequence was deleted (Fig. 1). When the deduced protein
structure of RANKL 2 was compared with that of RANKL 1, it was found
that RANKL 2 had a shorter intracellular domain than RANKL 1 (Fig. 2). The region from
Ser14 to Ala44 of RANKL 1
was deleted, and Ser14 and
Ala44 in RANKL 1 were converted to
Thr14 and Phe15 in RNAKL 2,
respectively. The transmembrane domain of RANKL 2 remained as it was.
RANKL 3 did not have the sequence encoding the transmembrane domain. In
addition, the frame of the codon was shifted at the junctional site
(nucleotide A28 and G29 of
the RANKL 3 sequence) in the RANKL 3 sequence, and a possible
initiation site of translation in RANKL 3 was Met118 (Fig. 2). Deduced amino acid sequences downstream from the transmembrane
domain of RANKL 1 and RANKL 2 were identical, and amino acids encoding
in the RANKL 3 sequence were shared in all three RANKL isoforms (Fig. 2).
|
|
), even though the expression of
RANKL mRNA in MC3T3-E1 cells was of a very low level and was detected
only in the cells treated with
1,25(OH)2D3 and Dex (Fig. 3). Using three primer sets that specifically detect RANKL 1, RANKL 2,
and RANKL 3, respectively (Fig. 1
), isoform-specific expression was
analyzed in the calvaria, thymus, lung, testis, ST2 cells, MC3T3-E1
cells, and thymocytes by RT-PCR. Under the conditions, all RANKL
isoforms were detected in these organs and cells (Fig. 4). However, in ST2 and MC3T3-E1 cells
treated with 1,25(OH)2D3
and Dex, the expression of RANKL 2 isoform was not detected in these
RT-PCR analyses (Fig. 4). To analyze the expression of the three RANKL
isoforms in thymocytes, T cell subsets were isolated from the thymus of
4-week-old mice. The immature double-negative
CD4-CD8- population, a
more mature double-positive
CD4+CD8+ population, and
mature single-positive
CD4+CD8- and
CD4-DC8+ populations were
isolated by cell sorting with a flow cytometer. RT-PCR analyses of
these populations indicated that all three RANKL isoforms were present
in CD4-CD8-,
CD4+CD8-, and
CD4-CD8+ cells. However,
in CD4+CD8+ cells, RANKL 2
was not detected (Fig. 4). Each PCR product was sequenced to confirm
that the real RANKL isoforms were amplified (data not shown). All
RT-PCR analyses were repeated three times or more, and the same results
were obtained. Murine glyceraldehyde-3-phosphate dehydrogenase (GAPDH)
was amplified for 25 cycles using one-tenth the amount of each
template cDNA as an internal control (Fig. 4).
|
|
, clones with a variety of expression
levels of each transgene were seen. In this experiment, clones with
relatively high levels of expression, clones RANKL 1-2 and 1-4, RANKL
2-1 and 2-3, and RANKL 3-1 and 3-2 were selected for further analyses.
A clone transfected with pMIKHyg B vector, without a cDNA insert (HV),
was selected as the control, and it was confirmed that the expression
of RANKL mRNA was not detected in this clone (Fig. 5). In addition, no
RANKL fragment was detected in the control clone by RT-PCR (data
not shown). The protein production in these clones was assessed by
immunofluorescence microscopy using a confocal laser scanning
microscope. Protein of RANKL 1 isoform was detected not only in the
endoplasmic reticulum and Golgi networks but also in the cytoplasmic
and membrane regions of each transfectant and looked predominantly
vesicularly distributed (Fig. 6C).
Protein of RANKL 2 isoform was predominantly detected in the
endoplasmic reticulum and Golgi networks (Fig. 6E). RANKL 3 protein was
diffusely seen in the cytoplasmic region of the cells and in the
endoplasmic reticulum and Golgi networks (Fig. 6G). The same results
were obtained in both clones selected for each RANKL transfectant. In
the control clone, which was transfected with pMIKHyg B vector without
a cDNA insert, no positive signal was detected (Fig. 6A). In each
transfectant treated with the preimmune serum, no positive signal was
detected (Fig. 6; B, D, F, and H).
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Nucleotides downstream from G226 in RANKL
1, G139 in RANKL 2, and G29
in RANKL 3 isoforms were completely conserved (Fig. 1 and 2).
Therefore, the extracellular domains of all 3 isoforms were almost
identical and were thought to have the capacity to bind with the
receptor RANK or the decoy receptor OPG. In addition, 28 nucleotides in
the N terminus regions of these 3 RANKL isoforms were also identical
(Fig. 1). Consequently, these 3 RANKL isoforms were suspected to be
produced by alternative splicing from the same gene. The
opgl-deficient mouse was produced by deleting nucleotides
405–501 in the RANKL 1 sequence, and the deletion occurred in a region
common to all 3 RANKL isoforms (16). In the
opgl-deficient mice, expression of RANKL was not detected by
Northern analysis using a full-length opgl cDNA probe
(16). It was confirmed that RANKL 1, RANKL 2, and RANKL 3
were produced from the same gene. Interestingly, treatment of ST2 and
MC3T3-E1 cells with
1,25(OH)2D3 and Dex greatly
up-regulated the expression of RANKL mRNA, as determined by Northern
analysis using a probe that detects all RANKL isoforms (Fig. 3), but
treatment with these steroids suppressed the expression of the
RANKL 2 isoform in RT-PCR analyses (Fig. 4). To conclude that the
expression of RANKL 2 is differentially regulated, more strict assay
should be used, but these results raise the possibility that RANKL 2 is
controlled by a different promoter from that of RANKL 1 and/or RANKL 3,
consistent with reports that 2 isoforms produced by alternative
splicing are controlled by 2 distinct promoters (22, 23, 24, 25).
The functional difference between the 2 membrane-associated isoforms,
RANKL 1 and RANKL 2, is unclear. However, a difference in the
intracellular domain between RANKL 1 and RANKL 2 causes the different
distributions of these proteins (Fig. 6, C and E), and this might
affect the function of these proteins as reported previously (26, 27). To clarify this point, further biological experiments,
using transfectants of each RANKL isoform, are needed.
In opgl-deficient mice, thymocyte development was impaired in the process of differentiation from CD4-CD8- double-negative CD44-CD25+ precursors to CD44-CD25- thymocytes (15), and the data indicated that OPGL was essential for development of thymocytes. It is of interest that immature CD4-CD8- thymocytes and mature CD4+CD8- and CD4-CD8+ thymocytes express all three RANKL isoforms, but CD4+CD8+ thymocytes do not express RANKL 2. The absence of RANKL 2 in CD4+CD8+ thymocytes suggests a specific function of this isoform. Many of the CD4+CD8+ thymocytes undergo apoptosis during the process of positive and negative selection before differentiation into mature CD4+CD8- or CD4-CD8+ thymocytes. The RANKL-RANK system is known to regulate the survival of dendritic cells (1, 2), and dynamic change in the expression of RANKL 2 during thymocyte development might associate with regulation of survival of thymocytes within the thymic microenvironment during the selection process.
In this study, we found two new isoforms of RANKL. One isoform, RANKL 3, did not have a transmembrane domain and was thought to be a soluble form. The other new isoform, RANKL 2, had a transmembrane domain, but its intracellular domain was much shorter than that of the original RANKL, RANKL 1. This structural difference caused the different distribution of these proteins and was suspected of causing the functional difference. Also of interest was the difference in the expression between RANKL 1/RANKL 3 and RANKL 2 isoforms after stimulation of ST2 and MC3T3-E1 cells by 1,25(OH)2D3 and Dex. Although the mechanism needs to be studied more closely, the results suggested that RANKL 2 was regulated by a promoter different from that of the other isoforms, and that the physiological functions of RANKL 1 and RANKL 2 were different. In addition, regulation of the ratio of these three RANKL isoforms might be important to the physiological function of the RANKL-RANK system in osteoclastogenesis and T cell differentiation within the thymic microenvironment.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received July 31, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(TNF-)-converting enzyme-like
protease in shedding of TRANCE, a TNF family member involved in
osteoclastogenesis and dendritic cell survival. J Biol Chem 274:13613–13618This article has been cited by other articles:
 |
|
 |
|
  G.E. Wise and G.J. King Mechanisms of Tooth Eruption and Orthodontic Tooth Movement J. Dent. Res., May 1, 2008; 87(5): 414 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. E. Kearns, S. Khosla, and P. J. Kostenuik Receptor Activator of Nuclear Factor {kappa}B Ligand and Osteoprotegerin Regulation of Bone Remodeling in Health and Disease Endocr. Rev., April 1, 2008; 29(2): 155 - 192. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Ukai, H. Yumoto, F. C. Gibson III, and C. A. Genco Macrophage-Elicited Osteoclastogenesis in Response to Bacterial Stimulation Requires Toll-Like Receptor 2-Dependent Tumor Necrosis Factor-Alpha Production Infect. Immun., February 1, 2008; 76(2): 812 - 819. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-M. Koh, Y.-S. Lee, C.-H. Byun, E.-J. Chang, H. Kim, Y. H. Kim, H.-H. Kim, and G. S. Kim {alpha}-Lipoic acid suppresses osteoclastogenesis despite increasing the receptor activator of nuclear factor {kappa}B ligand/osteoprotegerin ratio in human bone marrow stromal cells J. Endocrinol., June 1, 2005; 185(3): 401 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Roccisana, N. Kawanabe, H. Kajiya, M. Koide, G. D. Roodman, and S. V. Reddy Functional Role for Heat Shock Factors in the Transcriptional Regulation of Human RANK Ligand Gene Expression in Stromal/Osteoblast Cells J. Biol. Chem., March 12, 2004; 279(11): 10500 - 10507. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Heider, I. Zavrski, C. Jakob, C. Fleissner, O. Sezer, A. C. W. Zannettino, B. Pan, A. N. Farrugia, G. J. Atkins, and L. B. To Identification and Clinical Relevance of Receptor Activator of Nuclear Factor {kappa}B Ligand Expression of Myeloma Cells Cancer Res., January 15, 2004; 64(2): 773 - 775. [Full Text] [PDF]
|
|
|
|
 |
|
 U. Heider, C. Langelotz, C. Jakob, I. Zavrski, C. Fleissner, J. Eucker, K. Possinger, L. C. Hofbauer, and O. Sezer Expression of Receptor Activator of Nuclear Factor {kappa}B Ligand on Bone Marrow Plasma Cells Correlates with Osteolytic Bone Disease in Patients with Multiple Myeloma Clin. Cancer Res., April 1, 2003; 9(4): 1436 - 1440. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
