Cloning, Sequencing, and Functional ... - Wiley Online Library

JOURNAL OF BONE AND MINERAL RESEARCH Volume 15, Number 11, 2000 © 2000 American Society for Bone and Mineral Research

Cloning, Sequencing, and Functional Characterization of the Rat Homologue of Receptor Activator of NF-␬B Ligand* JIAKE XU,1 JAMIE WY TAN,1,2 LIN HUANG,1,2 XIU-HUI GAO,1 REBECCA LAIRD,1 DAN LIU,1 STAN WYSOCKI,1 and MING H. ZHENG1

ABSTRACT A complementary DNA (cDNA) encoding the rat homologue of receptor activator of NF-␬B ligand/ osteoprotegerin ligand/osteoclast differentiation factor/tumor necrosis factor (TNF)–related activationinduced cytokine (RANKL/OPGL/ODF/TRANCE) was cloned and sequenced from tibias of ovariectomized (OVX) rats. The predicted amino acid sequence of rat RANKL (rRANKL) has 84% and 96% identity to that of human and mouse RANKL, respectively, and 35% and 37% similarity to that of human and mouse TNF-related apoptosis-inducing ligand (TRAIL), respectively. RANKL transcripts were expressed abundantly in the thymus and bone tissues of OVX rats. rRANKL has a single hydrophobic region between residues 53 and 69, which is most likely to serve as a transmembrane domain. The long C-terminal region containing ␤-sheet–forming sequences of the TNF-like core is considered the extracellular region. Three truncated domains within the TNF-like core region were expressed as glutathione S-transferase (GST) fusion proteins and investigated for their ability to induce osteoclastogenesis. The results showed that GST-rRANKL (aa160 –318) containing the full TNF-like core region had the highest capability to induce the formation of osteoclast-like cells from RAW264.7 cells. GST-rRANKL (aa239 –318 and aa160 –268) had lesser degrees of osteoclast inductivity. Furthermore, the GST-rRANKL (aa160 –318) is capable of (1) inducing osteoclast formation from rat spleen cells in the presence of macrophage colony–stimulating factor (M-CSF), (2) stimulating mature rat osteoclast polarization and bone resorption ex vivo, and (3) inducing systemic hypercalcemia in vivo; thus the full TNF-like core region of rRANKL is an important regulator of calcium homeostasis and osteoclastic function. (J Bone Miner Res 2000;15:2178 –2186) Key words:

osteoclast, receptor activator of NK-␬B/osteoprotegerin ligand/osteoclast differentiation factor/ tumor necrosis factor–related activation-induced cytokine, bone resorption, hypercalcemia

INTRODUCTION ONE LOSS in ovariectomized (OVX) rats is analogous to postmenopausal osteoporosis in humans.(1) The characteristic changes in OVX rats include an early phase of rapid

B

*The sequence data has been submitted to the GenBank databases under access number AF187319.

bone loss followed by a slower loss, a high bone turnover rate with osteoclastic bone resorption greater than osteoblastic bone formation, and a greater loss of trabecular bone than cortical bone.(1–3) Our previous studies showed that OVX rats had enhanced levels of tartrate-resistant acid phosphatase (TRAP) and carbonic anhydrase II (CAII) messenger RNA (mRNA) in bone.(4) The degree of induction of TRAP and CAII mRNA was greater in the early phase (less

1 Department of Orthopedic Surgery, The University of Western Australia, WA Institute for Medical Research, QEII Medical Center, Nedlands, Western Australia, Australia. 2 These authors contributed equally to this work.

2178

CLONING OF RAT RANKL

than 2 weeks) than in the later phase (8 weeks) after ovariectomy.(4) It appears that the induction of TRAP and CAII mRNA was a result of increased osteoclastogenesis and osteoclastic bone resorption.(4,5) However, the molecular mechanism by which ovariectomy rapidly induced osteoclastogenesis and bone resorption is not fully understood. A recently discovered member of the tumor necrosis factor (TNF) receptor ligand family,(6) receptor activator of NF-␬B ligand (RANKL),(7) also called osteoclast differentiation factor (ODF),(8) osteoprotegerin ligand (OPGL),(9) and TNF-related activation-induced cytokine (TRANCE),(10) appears to be involved directly in the differentiation of monocytes/macrophages into osteoclasts.(8,9) Both membrane-bound and soluble forms of RANKL are able to induce osteoclast formation in vitro.(8,9) Mice with a disrupted RANKL gene show lack of osteoclasts, severe osteopetrosis, and a defect in tooth eruption, indicating that RANKL is essential for osteoclastogenesis in vivo.(11) In addition to its role in osteoclast function, RANKL has been shown to be a regulator of interactions between T cells and dendritic cells in vitro(10 –12) and of lymph node organogenesis and lymphocyte development in vivo.7 RANKLinduced osteoclastogenesis is mediated through the receptor RANK, which is expressed on the cell surface of osteoclast progenitors.(13,14) Transgenic mice expressing a soluble RANK-Fc fusion protein have a reduction in osteoclasts and display severe osteopetrosis.(13) The activity of RANKLinduced osteoclastogenesis is negatively regulated by a secreted TNF receptor–related molecule, OPG or osteoclastogenesis inhibitory factor (OCIF).(15–17) To further investigate the molecular mechanisms underlying bone loss, we have cloned a full-length complementary DNA (cDNA) encoding the rat homologue of RANKL from tibias of OVX rats. It was found that RANKL transcripts were expressed abundantly in thymus and bone tissues from OVX rats. A recombinant glutathione S-transferase (GST) fusion protein expressing the TNF-like core region was able to induce osteoclast formation in primary cultured cells from rat spleen in the presence of macrophage colony–stimulating factor (M-CSF) and in RAW cells in the absence of M-CSF. Administration of rat RANKL (rRANKL) in vivo induced the polarization of osteoclasts and systemic hypercalcemia in rats. Histomorphometric analysis showed that rRANKL-treated rats exhibited an increased bone resorbing activity in vivo.

MATERIALS AND METHODS Cloning and sequencing of rRANKL cDNA To isolate an rRANKL cDNA encoding the functional domain, the following degenerate oligonucleotide primers were designed based on the published consensus amino acid sequence (YFRAZM and AFKVR/QDID) of human and mouse RANKL/ODF/OPGL/TRANCE(8,9): RANKL sense, 5⬘TA(C/T)TT(C/T)(A/C)G(G/A/T/C)GC(G/A/T/C)CA(A/G)ATG3⬘; antisense, 5⬘(A/G)TC(A/T/G)AT(A/G)TC(A/G/T/C)(C/T)G(G/A/T/C)AC(C/T)TT(A/G)AA(G/A/T/ C)GC. Total RNA was isolated from Sprague–Dawley (S/D) rat bones 2 weeks after ovariectomy as previously described.(4) The reverse-transcription polymerase chain re-

2179

action (RT-PCR) products at the predicted size of 747 base pairs (bp) were gel-purified and cloned into the pCR2.1 T/A cloning vector (Invitrogen, Carlsbad, CA, U.S.A.). Clones containing the predicted size of PCR product were selected for automatic DNA sequencing and one clone, which shows striking homology with mouse and human RANKL, was identified. 3⬘- and 5⬘-end rapid amplification of cDNA ends (RACE) strategies were carried out using published procedures from Frohman et al.(18) to obtain the full length of the rRANKL coding sequence.

Northern blot analysis and RT-PCR To determine the tissue distribution of RANKL in OVX rats and normal rats, total RNA was isolated from various snaps frozen tissues using RNAzol solution according to the manufacturer’s instructions (Ambion, Inc., Austin, TX, U.S.A.). For Northern blot analysis, 15 ␮g of the total RNA were separated on a 1.2% agarose gel and transferred to Hybond-N⫹ membrane (Amersham International). The membrane was hybridized for 15 h at 42°C with a 32Plabeled rRANKL cDNA probe. As an internal control, the membrane was rehybridized with a 32P-labeled cDNA probe for the housekeeping gene 36B4.(19) For RT-PCR, singlestranded cDNA was prepared from 2 ␮g of total RNA using reverse transcriptase with an oligo-dT primer. Two microliters of each cDNA were subjected to 30 cycles of PCR (94°C, 40 s; 54°C, 40 s; and 72°C, 40 s) using rRANKLspecific primers: RANKL sense (5⬘CTTTGGATCCTAACAGAATATCAG3⬘) and RANKL antisense (5⬘AGGCTTCAGTCTATGTCTTGAACTTT3⬘). As an internal control, the single-stranded cDNA was PCR-amplified for 25 cycles using 36B4 primers sense (5⬘TCATTGTGGGAGCAGACA3⬘) and antisense (5⬘TCCTCCGACTCTTCCTTT3⬘).

Expression and purification of recombinant rRANKL To determine whether the functional region of rRANKL is capable of inducing osteoclastogenesis, cDNA fragments encoding the peptide sequences delineated by amino acids aa160 –318, aa239 –318, and aa160 –268 from the TNF-like core were cloned into the bacterial expression vectors pGEX-3X and pGEX-2T.(20) Correct insertions of RANKL cDNA in frame with GST were confirmed by DNA sequence analysis and the resultant plasmids were named p3rRANKL1 (aa160 –318), p3rRANKL2 (aa239 –318), and p3rRANKL3 (aa160 –268). To express GST fusion proteins, plasmids were transformed into the bacterial strain Top 10⬘. After growth in Luria-Bertani medium (LB) containing 100 ␮g/ml of ampicillin for 3 h at 30°C, isopropylthio-␤-D-galactoside (IPTG) was added to a final concentration of 0.1 mM and the bacterial culture was incubated further for 4 h at 30°C. Bacteria was harvested and lysed in a buffer containing 150 mM NaCl, 20 mM Tris-HCl, and 1 mM EDTA. The bacterial lysate obtained after addition of Triton X-100 to a final concentration of 1% was sonicated and centrifuged at 10,000 rpm for 10 minutes at 4°C. One milliliter of 50% of glutathione agarose beads (Amersham Pharmacia Biotech, Sydney, Australia) was added per liter of supernatant. After incubation for 1 h on ice, glutathione agarose beads were washed in phosphate-buffered saline

2180

(PBS), pH 7.2, containing 0.1% of Triton X-100 until no more protein was present in the washing solution monitored by spectrophotometry. GST-RANKL fusion protein was eluted with an increasing concentration of 10, 20, and 30 mM of reduced glutathione in 50 mM Tris-HCl (pH 8.0), the eluted proteins being analyzed on sodium dodecyl sulfate– polyacrylamide gel electrophoresis (SDS-PAGE) to estimate purity. Fractions with GST-RANKL greater than 95% pure were pooled and dialyzed with 2 liters of PBS overnight with one change. Protein concentration was determined on an SDSPAGE using a standard curve of serially diluted bovine serum albumin and by spectrophotometry.

In vitro osteoclastogenesis assays Cells harvested from the spleen of 2-day-old S/D rats were seeded in a 24-well plate to a density of 3 ⫻ 106 cells/well and cultured for 7–10 days in ␣-modified essential medium (␣-MEM) containing 10% fetal calf serum (FCS) and 10 ng/ml of M-CSF in the presence or absence of various concentrations of rRANKL.(21) The cells were then fixed with 4% paraformaldehyde in PBS for 10 minutes at room temperature and washed four times with PBS. The fixed cells were stained for TRAP using the Diagnostic Acid Phosphatase kit (Sigma, New South Wales, Australia) according to manufacturer’s instructions. A duplicate set of cells was stained for calcitonin receptor (CTR) using rabbit anti-rat/mouse CTR polyclonal antibody (provided by Dr. P.M. Sexton at Department of Pharmacology, the University of Melbourne, Victoria, Australia). For the amplification of rat CTR,(22) primer rCR1␣ sense 5⬘TTCCAGGGATTCTTTGTCGC3⬘ and primer rCR1␣ antisense 5⬘GCGGATGAGTCTTGCTGGA3⬘ were used and PCR was carried out with 30 cycles (94°C, 40 s; 62°C, 40 s; and 72°C, 40 s). For the amplification of mouse CTR,(23) primer mCR1␣ sense 5⬘TGGTTGAGGTTGTGCCCA3⬘ and primer mCR1␣ antisense 5⬘CTCGTGGGTTTGCCTCATC3⬘ were used (30 cycles, 94°C, 40 s; 62°C, 40 s; and 72°C, 40 s). Bone resorption pit assays were conducted by culture of spleen cells on bone slices in the presence of M-CSF and rRANKL and viewing of pits using the scanning electron microscope as previously described.(24) Mouse macrophage RAW264.7 cells also were used to test the inductive effect of rRANKL on osteoclastogenesis.(13) In brief, RAW264.7 cells at a density of 1 ⫻ 104 cells/well in a 24-well plate were cultured either in medium with GST (30 ng/ml) or medium with GST-rRANKL (aa160 –318; 30 ng/ml for polypeptides aa160 –318, aa239 –318, and aa160 – 268). All cultures were fed every 3 days by replacing used medium with respective fresh medium. After 10 days, all the cultures were fixed and proceeded to TRAP histochemistry, CTR immunohistochemistry,(25) and bone resorption pit assay to confirm the identity of osteoclasts. In both spleen and RAW264.7 cell cultures, TRAP⫹ multinuclear cells with more than three nuclei were scored and data were statistically analyzed by Student’s t-tests.

RANKL treatment of newborn rats Two-day-old S/D rats (n ⫽ 6 per group) were injected by the subcutaneous route with GST or GST-rRANKL

XU ET AL.

(aa160 –318) at a dose of 30 ␮g per animal. After 24 h, serum from individual rats was taken for the measurement of calcium. Tibias were removed and imprinted on glass slides as previously described.(24) The cells were stained for TRAP activity, permeabilized and incubated with rhodamine-phalloidin (Molecular Probes, Eugene, OR, U.S.A.) as previously described.(26) TRAP expressing multinucleated cells was identified and F-actin was visualized by confocal microscopy. The total numbers of F-actin ring forming osteoclasts (polarized osteoclasts) and non-Factin ring forming osteoclasts (nonpolarized osteoclasts ex vivo) were calculated under the fluorescence microscope.

RESULTS Cloning and sequencing of cDNA encoding rRANKL To isolate a cDNA fragment of rRANKL, degenerate oligonucleotide primers corresponding to the extracellular region of RANKL were designed based on the published consensus peptide sequences (YFRAQM and AFKVR/ QDID) from both human and mouse RANKL.(8,9) Since induction of osteoclastogenesis and bone resorption were very obvious at the early stage of ovariectomy, it is possible that RANKL may be up-regulated in OVX animals. To this end, total RNA extracted from tibias of 2-week-old OVX rats was used as a template for RT-PCR cloning of RANKL. The resultant RT-PCR products corresponding to the predicted size of 747 bp were gel-purified, subcloned into the pCR2.1 T/A cloning vector, and sequenced. One clone with striking sequence homology with human and mouse RANKL was identified after sequence analysis and homology search of the Genbank DNA database. To obtain the full length of rRANKL, 5⬘- and 3⬘-end RACE strategies were employed.(18) The predicted sequences of the rRANKL open reading frame are 318 amino acids in length with a single hydrophobic region between residues 53 and 69 (Fig. 1A). In addition, it has two potential N-linked glycosylation (NXT/S) sites located at residues 199 –202 (NMTL) and 264 –267 (NWSG), pointing out the characteristics of a type II transmembrane glycoprotein. Alignment of the deduced amino acid sequences shows that rRANKL has 84% identity to human RANKL and 96% identity to mouse RANKL (Fig. 1A). Both the rat and the mouse RANKLs appear to have a consensus TNF-␣ convertase (TACE) cleavage site (VGPQR/FSG).(27) In addition, rRANKL has 35% similarity to human and 37% to mouse TNF-related apoptosisinducing ligand (TRAIL)(28) and a lesser degree of similarity to other TNF-like molecules such as CD40 and Fas ligands. The strongest sequence similarity appears to be at their carboxyl regions (Fig. 1B). Taken together, rRANKL is a type II transmembrane glycoprotein, in which the C-terminal extracellular domain shows homology to other TNF family members.

Tissue distribution of RANKL expression To examine the tissue distribution of rRANKL mRNA transcripts, both Northern blot analysis and RT-PCR were used. A single RANKL transcript with an approximate size of 2.4 kb was detected in thymus and vertebrae from OVX


2181

FIG. 1. (A) Alignment of the predicted amino acid sequences of rRANKL with those of human and mouse RANKL. The predicted transmembrane regions are underlined. The metalloprotease disintegrin TACE cleavage site is marked by an arrow head. Right arrow marks the beginning of the TNF-like core domain within the extracellular region extending to the C terminus. (B) Alignment of the amino acid sequences of the rRANKL with those of human and mouse TRAIL. Identical amino acids are marked with black block and asteroids on the consensus lines. Similar amino acids are indicated with gray color and dots on the consensus lines.

2182

XU ET AL.

FIG. 2. Expression of RANKL in OVX rat tissues. (A) Northern blot analysis of total RNA from OVX rat tissues. Membranes were hybridized sequentially with P32-labeled cDNAs, rRANKL (upper panel), and 36B4 (lower panel). (B) RT-PCR analysis of RANKL (upper panel) and 36B4 (lower panel) in various OVX rat tissues. cDNA was synthesized using 2 ␮g of total RNA with oligo-dT and then subjected to PCR amplification using specific primers to rRANKL or 36B4. PCR products were separated on a 1.2% of agarose gel and Southern blotting was performed using P32-labeled cDNAs, rRANKL, or 36B4.

rats on Northern blot analysis (Fig. 2A). The transcript was also observed in tibias and very weakly in spleen and lung from these rats (data not shown). Other tissues including liver and kidney did not express detectable RANKL mRNA using Northern blot methodology (Fig. 2A), even when polyA mRNA was used. Because previous studies had shown that additional tissues in mouse and human express ODF/RANKL/ OPGL/TRANCE,(8 –10) we employed RT-PCR with rRANKLspecific primers followed by Southern blot hybridization with the rRANKL cDNA probe to detect weaker levels of expression. As shown in Fig. 2B, expression of RANKL mRNA in OVX rats was most abundant in thymus, followed by tibia, vertebrae, and rib crest, and then spleen and much lower levels in lung, with just traces of expression in liver and kidney. RANKL transcripts were undetectable in brain and heart. A similar pattern of RANKL mRNA expression was observed in normal sham-operated rats where the level of RANKL mRNA expression was also highest in thymus, followed by bone tissues, spleen, and lung (data not shown). There was no obvious difference in the levels of RANKL mRNA between OVX and sham-operated rats (data not shown).

TNF-like core region is the functional domain of rRANKL rRANKL has a single hydrophobic region between residues 53 and 69, which is most likely to serve as a transmembrane domain. The long C-terminal extracellular region containing regions of ␤-sheet–forming sequences in the TNF-like core (Fig. 3A) is considered to be the extracellular region. To determine the active ligand domain at the C-terminal region, three sequence domains within the TNF-like core region termed p3rRANKL1, p3rRANKL2, and p3rRANKL3 (Fig.

3B) were selected and cloned into the bacterial expression vectors pGEX-3X and pGEX-2T. Soluble forms of recombinant rRANKL were readily affinity-purified to greater than 95% purity (Fig. 3C). The biological activities of various truncated proteins were investigated using RAW264.7 (mouse monocyte-macrophage cell line; American Type Culture Collection, Rockville, MD, U.S.A.) cell cultures. As shown in Fig. 3D, GST-rRANKL (aa160 –318) containing the full TNF-like core region had the highest capability to induce the formation of osteoclast-like cells from RAW264.7 cells. GST-rRANKL (aa239 –318) and (aa160 –268) had lesser degrees of osteoclast inductivity, whereas GST alone had no effect on osteoclastogenesis. Osteoclast-like cells formed by the fusion of RAW264.7 cells are capable of resorbing bone and expressing CTR as detected by immunohistochemistry (data not shown). These results indicated that the active functional domain of rRANKL requires the complete TNF-like core region. To further examine the osteoclast inductivity of GST-rRANKL (aa160 –318), spleen cells isolated from 2-day-old S/D rats were treated with the recombinant proteins at various concentrations in the presence of M-CSF, for 7–10 days. As shown in Fig. 4, addition of recombinant GST-rRANKL (aa160 –318) induced the formation of TRAP⫹ multinuclear cells. These cells were positive for CTR, as shown by immunohistochemistry and RT-PCR, and were capable of resorbing bone. The optimal inductivity was observed at a concentration of 30 ng/ml.

rRANKL induces osteoclast polarization and hypercalcemia in vivo To determine the biological effects of rRANKL (aa160 – 318) in vivo, 2-day-old S/D rats were injected by the sub-


2183

FIG. 3. Effect of truncated proteins of the TNF-like core region on osteoclastogenesis. (A) Schematic diagram of the structure of rRANKL. C, cytoplasmic region; TM, transmembrane region; E, extracellular region; arrow head, the TACE cleavage site. (B) Truncated regions of RANKL cDNA cloned into the pGEX expression vectors. (C) Coomassie blue–stained polyacrylamide gel showing the induction and purification of rRANKL GST fusion proteins from Escherichia coli. (D) Inductive activity of osteoclastogenesis of truncated proteins for the TNF-like core region. Murine myeloid Raw264.7 cells were cultured in the presence of 30 ng/ml of various recombinant truncated proteins for 7 days and TRAP⫹ multinucleated cells were scored. Inductive activity was calculated as percentage of total numbers of TRAP⫹ cells for each truncated protein relative to the total numbers of TRAP⫹ cells for GST-rRANKL (aa160 –318).

cutaneous route either with 30 ␮g of GST-rRANKL (aa 160 –318) or with GST alone. After 24 h, sera were taken for the measurement of calcium levels. Long bone tissues were also harvested for histomorphometry and Rhodamine phalloidin and TRAP staining. The numbers of TRAP⫹ multinuclear cells ex vivo with or without the presence of F-actin rings were scored using confocal microscopy. As shown in Fig. 5A, injection of GST-rRANKL (aa160 –318) induced hypercalcemia in rats. The level of serum-ionized calcium in GST-rRANKL (aa160 –318)–injected animals was significantly greater than that in GST-injected ones. It was evident that the percentage of TRAP⫹ osteoclasts with an F-actin ring was significantly greater in rats that received GST-rRANKL (aa160 –318) than in those that received GST alone (Figs. 5B and 5C). These results indicated that rRANKL stimulates polarization of mature osteoclasts. Further histomorphometry analysis revealed that the total numbers of osteoclast resorbing surfaces in GST-rRANKL (aa160 –318)–injected rats were significantly greater than that in GST-injected rats. On the other hand, although the total numbers of osteoclasts per millimeter in GST-rRANKL (aa160 –318) seems higher than in GST-injected animals, there was no obvious statistical significance between them (Fig. 5D). Taken together, these

results indicated that the hypercalcemia induced after 24 h by the injection of rRANKL is most likely caused by the induction of osteoclastic bone resorption.

DISCUSSION Studies on the structure, function, regulation, and signaling pathways of RANKL/OPGL/ODF/TRANCE are attracting much attention because of the fact that RANKL is the key factor that is essential for osteoclast formation.(6 –11) Here, we have cloned the rat homologue of murie RANKL/ OPGL/ODF/TRANCE and extended investigation into the functional characterization of the TNF-like core domain of RANKL. We have shown rRANKL action on calcium homeostasis and osteoclastic cell function. RANKL/ODF/OPGL/TRANCE is a type II transmembrane glycoprotein with sequence homology to other TNF family members, most notably TRAIL. It has been shown that TRAIL binds OPG, a secreted TNF-like receptor for RANKL and therefore indirectly inhibits RANKL-induced osteoclastogenesis.(29) However, it remains uncertain whether RANKL and TRAIL bind to the same region of OPG. In this study, the

2184

FIG. 4. Effect of the TNF-like core region on osteoclast development from primary rat spleen cells. (A) Spleen cells were cultured in the presence of 10 ng/ml of M-CSF alone (a) or 10 ng/ml of M-CSF and 30 ng/ml of GST-rRANKL (aa160 –318) for 7 days (b). Multinucleated cells formed expressed CTR (c). Formation of bone resorption pits by rat spleen cell– derived osteoclasts also were observed after 10 days culture (d) (see Material and Methods section). (B) RTPCR analysis of CTR in rat spleen cell– developed osteoclasts. Total cellular RNA was extracted from spleen cells cultured in media alone or in media supplemented with 10 ng/ml M-CSF or 10 ng/ml M-CSF and 30 ng/ml of GST-rRANKL (aa160 – 318) and RT-PCR was performed using CTR-specific primers (upper) or 36B4 primers (lower). (C) Dose-dependent response of rRANKL on osteoclastogenesis. Spleen cells were cultured in the presence of 10 ng/ml of M-CSF and various doses (10 – 60 ng/ml) of GST-rRANKL (aa160 –318) for 8 days and stained for TRAP. Total numbers of TRAP⫹ multinucleated cells per 13-mm coverslips were scored.

TNF-like core domain of rRANKL was able to induce osteoclast formation in both rat spleen cell cultures and mouse RAW cells. Protein truncation analyses of the TNF-like core domain indicated that both RANKL aa239 –318 and aa160 – 268 have less potency in osteoclast induction in comparison

XU ET AL.

with the full-length TNF-like core domain. This result indicates that ligand binding of RANKL requires the complete TNF-like core sequences. The multinucleated cells induced by the TNF-like core domain are TRAP⫹ and capable of forming resorption pits on bone slices and therefore satisfy the major criteria characteristic of functional osteoclasts. Although RANKL is considered to be a membrane-bound protein so that the biological actions require cell-cell contact,(6 –10) a recent study by Lum et al.(27) has shown that mouse RANKL, like TNF-␣, can be shed by the metalloprotease disintegrin TACE, which can cleave the molecule at the stalk region resulting in the release of the soluble form of RANKL into the circulation. The soluble mouse RANKL has a potent effect on osteoclastogenesis.(27) Similarly, rRANKL also contained a TACE cleavage site at the stalk region (Fig. 1), indicating that RANKL may act as a soluble cytokine in OVX rats. Indeed, the existence of a soluble form of RANKL also has been suggested by Nagai.(30) RANKL expression, as detected by Northern blot analysis in the present study, was most abundant in thymus, and it is possible that thymusderived soluble RANKL may be an important factor in the rapid onset of ovariectomy-induced bone loss. In support of this hypothesis, a recent report has indicated that both systemic and local T cell activation can lead to RANKL production and subsequent bone loss in arthritis.(31) Furthermore, it has been shown that the expression of RANKL in T cells has been characterized as an immediate early response gene after T cell receptor activation,(10) although further study is required to determine the relationship of ovariectomy and RANKL production in activated T cells. Previous studies by Burgess et al.(32) and Lacey et al.(9) have shown that RANKL/OPGL/ODF/TRANCE injection caused elevation of ionized calcium in the blood. The increase in ionized calcium in the blood was evident as early as 1 h after the injection although frank hypercalcemia was not observed until 2 days. The direct relationship of increased ionized calcium in blood with osteoclast activation has not been clearly defined. Gut calcium absorption seems unlikely to be the explanation for increased blood calcium levels.(32) In this study we have examined the polarization of osteoclasts ex vivo in rats injected with rRANKL. The polarization of osteoclasts is critical for bone resorption, a process in which osteoclasts form a tight seal with the substratum resulting in an isolated resorption compartment.(33,34) We have extended the functional analysis of RANKL in vivo and have shown that the TNF-like core domain induces polarization of osteoclasts. The increase in ionized calcium in blood is coincident with the induction of osteoclast polarization. Furthermore, histomorphometric analysis also showed that rRANKL-injected rats displayed high numbers of bone resorbing areas per bone surface, whereas the total number of osteoclasts seems not to change significantly. Thus, based on the data, we suggest that RANKL-induced hypercalcemia is at least in part caused by the activation of existing osteoclasts at the bone surface (direct induction of osteoclast polarization). In summary, we have cloned and sequenced a rat homologue of RANKL/OPGL/ODF/TRANCE. Functional characterization showed that the TNF-like core domain of RANKL is essential for the induction of osteoclastogenesis and activation of mature osteoclasts. The TNF-like core


2185

FIG. 5. Induction of systemic hypercalcemia in vivo and osteoclast polarization ex vivo by the TNF-like core region of rRANKL. Two-day-old S/D rats (n ⫽ 6 per group) were injected by the subcutaneous route with GST-rRANKL (aa160 –318) or GST alone as controls at a dose of 30 ␮g per animal. After 24 h, serum from individual rats was taken for the measurement of calcium. (A) Comparison of serum calcium levels in newborn rats injected with either GST-rRANKL (aa160 –318) or GST alone (control). (B) F-actin ring formation in osteoclasts. TRAP⫹ osteoclasts in newborn rats injected with GST-rRANKL (aa160 –38) (a); osteoclasts under the same field as (a) display F-actin ring formation (b); TRAP⫹ osteoclasts in newborn rats injected with GST alone (c); osteoclasts under the same field as (c) do not display F-actin ring formation (d). (C) Comparison of percentages of polarized osteoclasts (F-actin ring) in newborn rats injected with either GST-rRANKL (aa160 –318) or GST (control). (D) Histomorphometry analysis of tibias in rats injected with either GST-rRANKL (aa160 –318) or GST. The total numbers of osteoclasts/bone perimeter and osteoclast resorbing surfaces/ bone perimeters were determined. #p ⬎ 0.05; *p ⬍ 0.05; **p ⬍ 0.01.

domain of RANKL is capable of inducing hypercalcemia and our evidence suggested that this might be caused by the activation of existing mature osteoclasts. Thus, the rat homologue of RANKL has an important role in regulation of calcium homeostasis and osteoclastic function.

ACKNOWLEDGMENTS This work was partly supported by grants from the National Health and Medical Research Council, Australian Research Council, the Orthopedic Research and Education Fund, and the Medical Research Fund of Western Australia awarded to M.H.Z.

REFERENCES 1. Kimmel DB 1996 Animal models for in vivo experimentation in osteoporosis research. In: Marcus R, Feldman D, Kelsey J

2. 3.

4.

5. 6.

7.

(eds.) Osteoporosis. Academic Press, San Diego, CA, U.S.A., pp. 671– 690. Kaln DN 1991 The ovariectomised rat model of post menopausal bone loss. Bone Miner 15:175–192. Wronski TJ, Cintron M, Daun LM 1988 Temporal relationship between bone loss and increased bone turnover in ovariectomised rats. Calcif Tissue Int 42:179 –183. Zheng MH, Lau T-TA, Prince R, Criddle A, Wysocki S, Beilharz M, Papadimitriou JM, Wood DJ 1995 17␤ estradial suppresses gene expression of tartrate-resistant acid phosphatase and carbonic anhydrase II in ovariectomised rats. Calcif Tissue Int 56:166 –169. Huang WH, Zheng MH 1999 Steroid hormones and bone. Histol Histopathol 14:1257–1268. Suda T, Takahashi N, Udagawa N, Jimi E, Gillespie MT, Martin TJ 1999 Modulation of osteoclast differentiation and function by the new members of the tumor necrosis factor receptor and ligand families. Endocr Rev 20:345–357. Anderson DM, Maraskovsky E, Billingsley WL, Dougall WC, Tometsko ME, Roux ER, Teepe MC, DuBose RF, Cosman D, Galibert L 1997 A homologue of the TNF receptor and its

2186

8.

9.

10.

11.

12.

13.

15.

16.

17.

18.

19.

20. 21.

ligand enhance T-cell growth and dendritic-cell function. Nature 390:175–179. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, Mochizuki S, Tomoyasu A, Yano K, Goto M, Murakami A, Tsuda E, Morinaga T, Higashio K, Udagawa N, Takahashi N, Suda T 1998 Osteoclast differentiation factor is a ligand for osteoprotegerin/ osteoclastogenesis-inhibitory factor and is identical to TRANCE/ RANKL. Proc Natl Acad Sci USA 95:3597–3602. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, Burgess T, Elliott R, Colombero A, Elliott G, Scully S, Hsu H, Sullivan J, Hawkins N, Davy E, Capparelli C, Eli A, Qian YX, Kaufman S, Sarosi I, Shalhoub V, Senaldi G, Guo J, Delaney J, Boyle WJ 1998 Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93:165–176. Wong BR, Rho J, Arron J, Robinson E, Orlinick J, Chao M, Kalachikov S, Cayani E, Bartlett FS III, Frankel WN, Lee SY, Choi Y 1997 TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells. J Biol Chem 272:25190 –25194. Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, Morony S, Oliveira-dos-Santos AJ, Van G, Itie A, Khoo W, Wakeham A, Dunstan CR, Lacey DL, Mak TW, Boyle WJ, Penninger JM 1999 OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397:315–323. Wong BR, Josien R, Lee SY, Sauter B, Li HL, Steinman RM, Choi Y 1997 TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor. J Exp Med 186:2075–2080. Hsu H, Lacey DL, Dunstan CR, Solovyev I, Colombero A, Timms E, Tan HL, Elliott G, Kelley MJ, Sarosi I, Wang L, Xia XZ, Elliott R, Chiu L, Black T, Scully S, Capparelli C, Morony S, Shimamoto G, Bass MB, Boyle WJ 1999 Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci USA 96:3540 –3545. Yasuda H, Shima N, Nakagawa N, Mochizuki SI, Yano K, Fujise N, Sato Y, Goto M, Yamaguchi K, Kuriyama M, Kanno T, Murakami A, Tsuda E, Morinaga T, Higashio K 1998 Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): A mechanism by which OPG/OCIF inhibits osteoclastogenesis in vitro. Endocrinology 139:1329 –1337. Mizuno A, Amizuka N, Irie K, Murakami A, Fujise N, Kanno T, Sato Y, Nakagawa N, Yasuda H, Mochizuki S, Gomibuchi T, Yano K, Shima N, Washida N, Tsuda E, Morinaga T, Higashio K, Ozawa H 1998 Severe osteoporosis in mice lacking osteoclastogenesis inhibitory factor/osteoprotegerin. Biochem Biophys Res Commun 247:610 – 615. Simonet WS, Lacey DL, Dunstan CR, Kelley M, Chang MS, Luthy R, Nguyen HQ, Wooden S, Bennett L, Boone T, Shimamoto G, DeRose M, Elliott R, Colombero A, Tan HL, Trail G, Sullivan J, Davy E, Bucay N, Renshaw-Gegg L, Hughes TM, Hill D, Pattison W, Campbell P, Sander S, Van G, Tarpley J, Derby P, Lee R, Amgen EST Program, Boyle WJ 1997 Osteoprotegerin: A novel secreted protein involved in the regulation of bone density. Cell 89:309 –319. Frohman MA, Dush MK, Martin GR 1988 Rapid production of full-length cDNAs from rare transcripts: Amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci USA 85:8998 –9002. Wool IG, Chan YL, Gluck A, Suzuki K 1991 The primary structure of rat ribosomal proteins P0, P1 and P2 and a proposal for a uniform nomenclature for mammalian and yeast ribosomal proteins. Biochimie 73:861– 870. Smith DB, Johnson KS 1988 Single-step purification of polypeptides expressed in Escherichia coli as fusions with glutathione S-transferase. Gene 67:31– 40. Udagawa N, Takahashi N, Akatsu T, Tanaka H, Sasaki T, Nishihara T, Koga T, Martin TJ, Suda T 1990 Origin of

XU ET AL.

22. 23.

24. 25.

26.

27.

28.

29.

30.

31.

32.

33. 34.

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. Albrandt K, Mull E, Brady EM, Herich J, Moore CX, Beaumont K 1993 Molecular cloning of two receptors from rat brain with high affinity for salmon calcitonin. FEBS Lett 325:225–232. Yamin M, Gorn AH, Flannery MR, Jenkins NA, Gilbert DJ, Copeland NG, Tapp DR, Krane SM, Goldring SR 1994 Cloning and characterization of a mouse brain calcitonin receptor complementary deoxyribonucleic acid and mapping of the calcitonin receptor gene. Endocrinology 135:2635–2643. Zheng MH, Fan Y, Wysocki S, Wood DJ, Papadimitriou JM 1993 Detection of mRNA for carbonic anhydrase II in human osteoclastlike cells by in situ hybridization. J Bone Miner Res 8:113–118. Quinn JM, Morfis M, Lam MH, Elliott J, Kartsogiannis V, Williams ED, Gillespie MT, Martin TJ, Sexton PM 1999 Calcitonin receptor antibodies in the identification of osteoclasts. Bone 25:1– 8. Yovich S, Seydel U, Papadimitriou JM, Nicholson GC, Wood DJ, Zheng MH 1998 Evidence that failure of osteoid bone matrix resorption is caused by perturbation of osteoclast polarization. Histochem J 30:267–273. Lum L, Wong BR, Josien R, Becherer JD, ErdjumentBromage H, Schlondorff J, Tempst P, Choi Y, Blobel CP 1999 Evidence for a role of a tumor necrosis factor-alpha (TNF-alpha)converting enzyme-like protease in shedding of TRANCE, a TNF family member involved in osteoclastogenesis and dendritic cell survival. J Biol Chem 274:13613–13618. Wiley SR, Schooley K, Smolak PJ, Din WS, Huang CP, Nicholl JK, Sutherland GR, Smith TD, Rauch C, Smith CA, Goodwin RG 1995 Identification and characterization of a new member of the TNF family that induces apoptosis. Immunity 3:673– 682. Emery JG, McDonnell P, Burke MB, Deen KC, Lyn S, Silverman C, Dul E, Appelbaum ER, Eichman C, DiPrinzio R, Dodds RA, James IE, Rosenberg M, Lee JC, Young PR 1998 Osteoprotegerin is a receptor for the cytotoxic ligand TRAIL. J Biol Chem 273:14363–14367. Nagai M, Kyakumoto S, Sato N 2000 Cancer cells responsible for humoral hypercalcemia express mRNA encoding a secreted form of ODF/TRANCE that induces osteoclast formation. Biochem Biophys Res Commun 269:532–536. Kong YY, Feige U, Sarosi I, Bolon B, Tafuri A, Morony S, Capparelli C, Li J, Elliott R, McCabe S, Wong T, Campagnuolo G, Moran E, Bogoch ER, Van G, Nguyen LT, Ohashi PS, Lacey DL, Fish E, Boyle WJ, Penninger JM 1999 Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 402:304 –309. Burgess TL, Qian Y, Kaufman S, Ring BD, Van G, Capparelli C, Kelley M, Hsu H, Boyle WJ, Dunstan CR, Hu S, Lacey DL 1999 The ligand for osteoprotegerin (OPGL) directly activates mature osteoclasts. J Cell Biol 145:527–538. Lakkakorpi PT, Vaananen HK 1991 Kinetics of the osteoclast cytoskeleton during the resorption cycle in vitro. J Bone Miner Res 6:817– 826. Vaananen HK, Horton M 1995 The osteoclast clear zone is a specialized cell-extracellular matrix adhesion structure. J Cell Sci 108:2729 –2732.

Address reprint requests to: Ming H. Zheng, M.D, Ph.D., M.R.C.Path. Department of Orthopedic Surgery University of Western Australia Nedlands 6009 WA, Australia Received in original form February 14, 2000; in revised form April 24, 2000; accepted June 12, 2000.